RELATED APPLICATIONS

This application claims priority to U.S. Provisional Application Serial No. 63/378,853, filed October 07, 2022, entitled “IMMUNOSTIMULATORY BACTERIA FOR CONVERTING MACROPHAGES INTO A PHENOTYPE AMENABLE TO TREATMENT AND COMPANION DIAGNOSTIC,” to Applicant Actym Therapeutics, Inc., and inventors Christopher D. Thanos, Laura Hix Glickman, Chingnam Cheung, Alexandre Charles Michel lannello, Bret Nicholas Peterson, Nicholas Boyce Woodall, and Akshata Udyavar; to U.S. Provisional Application Serial No. 63/311,424, filed February 17, 2022, entitled “IMMUNO STIMULATORY BACTERIA FOR CONVERTING MACROPHAGES INTO A PHENOTYPE AMENABLE TO TREATMENT AND COMPANION DIAGNOSTIC,” to Applicant Actym Therapeutics, Inc., and inventors Christopher D. Thanos, Laura Hix Glickman, Chingnam Cheung, Alexandre Charles Michel lannello, Bret Nicholas Peterson and Nicholas Boyce Woodall; and to each of U.S. Provisional Application Serial Nos. 63/277,601 and 63/278,076, filed November 09, 2021, and November 10, 2021, respectively, each entitled “IMMUNOSTIMULATORY BACTERIA-BASED VACCINES, THERAPEUTICS, AND RNA DELIVERY PLATFORMS,” each to Applicant Actym Therapeutics, Inc., and inventors Laura Hix Glickman, Chingnam Cheung, Alexandre Charles Michel lannello, Bret Nicholas Peterson and Christopher D. Thanos.

This application is related to International PCT Application No. PCT/US2021/045832, filed August 12, 2021, and published as WO 2022/036159 on February 17, 2022, entitled “IMMUNOSTIMULATORY BACTERIA-BASED VACCINES, THERAPEUTICS AND RNA DELIVERY PLATFORMS,” to Applicant Actym Therapeutics, Inc., and inventors Laura Hix Glickman, Bret Nicholas Peterson, Haixing Kehoe, Alexandre Charles Michel lannello, and Christopher D. Thanos.

This application is related to U.S. Provisional Application Serial No. 63/064,869, filed on August 12, 2020, entitled “IMMUNOSTIMULATORY BACTERIA DELIVERY PLATFORM,” to Applicant Actym Therapeutics, Inc., and inventors Laura Hix Glickman, Christopher D. Thanos, Alexandre Charles Michel lannello, Chris Rae, and Haixing Kehoe.

This application is related to U.S. Provisional Application Serial No. 63/188,443, filed May 13, 2021, entitled “IMMUNO STIMULATORY BACTERIA DELIVERY PLATFORM,” to Applicant Actym Therapeutics, Inc., and inventors Laura Hix Glickman, Christopher D. Thanos, Alexandre Charles Michel lannello, Chris Rae, and Haixing Kehoe.

This application also is related to U.S. Application Serial No. 17/320,200, filed May 13, 2021, entitled “IMMUNOSTIMULATORY BACTERIA DELIVERY PLATFORMS AND THEIR USE FOR DELIVERY OF THERAPEUTIC PRODUCTS,” to Applicant Actym Therapeutics, Inc., and inventors Laura Hix Glickman, Christopher D. Thanos, Alexandre Charles Michel lannello, Chris Rae, Haixing Kehoe, Bret Nicholas Peterson, and Chingnam Cheung.

This application is related to International Patent Application No. PCT/US2020/060307, filed on November 12, 2020, and published as WO 2021/097144 on May 20, 2021, entitled “IMMUNO STIMULATORY BACTERIA DELIVERY PLATFORMS AND THEIR USE FOR DELIVERY OF THERAPEUTIC PRODUCTS,” to Applicant Actym Therapeutics, Inc., and inventors Christopher D. Thanos, Laura Hix Glickman, Alexandre Charles Michel lannello, Chris Rae, Haixing Kehoe, Bret Nicholas Peterson, and Chingnam Cheung.

This application is related to International Patent Application No. PCT/US2020/020240, filed on February 27, 2020, and published as WO 2020/176809 on September 03, 2020, and to co-pending U.S. Patent Application Serial No. 16/824,500, filed on March 19, 2020, and published as U.S. Publication No. US 2020/0270613 Al on August 27, 2020, each entitled “IMMUNOSTIMULATORY BACTERIA ENGINEERED TO COLONIZE TUMORS, TUMOR-RESIDENT IMMUNE CELLS, AND THE TUMOR MICROENVIRONMENT,” to Applicant Actym Therapeutics, Inc., and inventors Christopher D. Thanos, Laura Hix Glickman, Justin Skoble, Alexandre Charles Michel lannello, and Haixing Kehoe.

This application also is related to International Patent Application No. PCT/US2018/041713, filed on July 11, 2018, and published as WO 2019/014398 on January 17, 2019, and to U.S. Patent Application Serial No. 16/033,187, issued on November 09, 2021, as U.S. Patent No. 11,168,326, filed on July 11, 2018, and published as U.S. Publication No. 2019/0017050 Al on January 17, 2019, each entitled “ENGINEERED IMMUNOSTIMULATORY BACTERIAL STRAINS AND USES THEREOF,” and each to Actym Therapeutics, Inc., and inventors Christopher D. Thanos, Laura Hix Glickman, and Justin Skoble.

This application also is related to International Patent Application No. PCT/US2019/041489, filed on July 11, 2019, and published as WO 2020/014543, on January 16, 2020, entitled “ENGINEERED IMMUNO STIMULATORY BACTERIAL STRAINS AND USES THEREOF,” to Actym Therapeutics, Inc., and inventors Christopher D. Thanos, Laura Hix Glickman, Justin Skoble, and Alexandre Charles Michel lannello.

The immunostimulatory bacteria and methods provided in each of these applications can be modified and used as described in this application, and such bacteria are incorporated by reference herein. Where permitted, the subject matter of each of these applications is incorporated by reference in its entirety.

INCORPORATION BY REFERENCE OF SEQUENCE LISTING PROVIDED ELECTRONICALLY

An electronic version of the Sequence Listing is filed herewith, the contents of which are incorporated by reference in their entirety. The electronic file was created on November 08, 2022, is 2,127 kilobytes in size, and is titled 1710PCSEQ001.xml. FIELD OF THE INVENTION

Provided are attenuated immunostimulatory bacteria with genomes that are modified to, for example, reduce undesirable inflammatory responses and toxicity, and improve the anti-tumor activity and/or immunostimulatory activity by increasing resistance to complement inactivation, by reducing immune cell death, by promoting adaptive immunity, and by enhancing T-cell function. The increase in colonization of phagocytic cells, for anti-cancer uses improves the delivery of encoded therapeutic products to the tumor microenvironment and to tumors, and permits, among other routes, systemic administration of the immunostimulatory bacteria. Immunostimulatory bacteria provided herein bacterioinfect proliferating macrophages, which upon expression of their encoded payload, result in a heretofore undescribed M1/M2 hybrid phenotype.

BACKGROUND

The field of cancer immunotherapy has made great strides, as evidenced by the clinical successes of anti-CTLA-4, anti-PD-1 and anti-PD-Ll immune checkpoint antibodies (see, e.g., Buchbinder et al. (2015) J. Clin. Invest. 125:3377-3383; Hodi et al. (2010) N. Engl. J. Med. 363(8):711-723; and Chen et al. (2015) J. Clin. Invest. 125:3384-3391). Tumors have evolved a profoundly immunosuppressive environment. They initiate multiple mechanisms to evade immune surveillance, reprogram anti-tumor immune cells to suppress immunity, and continually mutate resistance to the latest cancer therapies (see, e.g., Mahoney et al. (2015) Nat. Rev. Drug Discov. 14(8):561-584). Designing immunotherapies and cancer therapies that overcome immune tolerance and escape, while limiting the autoimmune-related toxicities of current immunotherapies, challenges the field of immuno-oncology. Hence, additional and innovative immunotherapies and other therapies are needed. SUMMARY

Immunostimulatory bacteria include other gram-negative Enterobacteriaceae, and gram-positive bacteria, such as Listeria and Shigella species. As described herein bacteria can accumulate in immunoprivileged and/or immunosuppressed cells and tissues, and hence have been used for delivery of active molecules, such as therapeutic molecules, to such cells and tissues, which include tumors, and the tumor microenvironment, and immune cells, such as phagocytic cells, including macrophages. As described herein, genome modifications can improve such properties of bacteria. For example, the bacteria can be modified so that they do not infect epithelial cells, but retain or have increased infection or uptake by phagocytic cells, such as macrophages. It also is shown herein, that for delivery of active molecules, such as proteins and nucleic acids, it is advantageous if the macrophages are proliferating, particularly for expression of nucleic acids, which requires entry across the nuclear membrane for transcription in the nuclei. The active molecules can be encoded on plasmids under control of transcription and/or translation regulatory sequences for expression. The encoding nucleic acid can be under control of prokaryotic regulatory sequences for expression in the bacteria for delivery of proteins; or can be under the control of bacterial promoters for transcription, but can be encoded in nucleic acid such that the transcripts are not translated in the bacterial hosts so that RNA is delivered to the cells and tissues; or the encoding nucleic acids can be under control of eukaryotic regulatory sequences, such as eukaryotic promoters, so that plasmids are delivered to the cells and tissues. As result, the immunostimulatory bacteria herein have a variety of applications, including, but not limited to, anti-tumor therapeutics, vaccines, including anti-cancer vaccines, and

- 4 -

RECTIFIED SHEET (RULE 91) ISA/EP vaccines against infective agents, for RNA delivery, for protein delivery, and other applications apparent from the description and examples herein.

Immunostimulatory Bacteria and Proliferating Macrophages

Tumors have evolved a profoundly immunosuppressive environment. They initiate multiple mechanisms to evade immune surveillance, reprogram anti-tumor immune cells to suppress immunity, and continually mutate resistance to the latest cancer therapies (see, e.g., Mahoney et al. ( QiS) Nat. Rev. Drug Discov. 14(8):561- 584). Across the solid tumor spectrum only inflamed tumors, rich in exhausted T- cells, respond well to checkpoint immunotherapies (such as non-small cell lung cancer (NSCLC) and melanoma). Tumors that have T-cells that are excluded to the tumor stroma (T-cell excluded), and tumors without T-cells (immune desert), do not respond to existing immunotherapies, and, thus, pose an unmet therapeutic need across solid tumors. Immunotherapy effectiveness, such as immune checkpoint inhibitor-mediated antitumor responses, depend on the infiltration of T cells capable of recognizing and killing tumor cells. These immunotherapies are not effective in so- called "cold tumors," or immune desert or T-cell excluded tumors, which are characterized by the lack of T-cell infiltration. The tumor core of T-cell excluded and of immune desert tumors is rich in myeloid cells. T-cell excluded tumors have an abundance of tumor-associated macrophages (TAMs) that are immunosuppressive and form a barrier that keeps the T-cells out (Keren et al. (2018) Cell 174:1373-1387; Bindea et al. (2013) Immunity 39:782-795).

Successful attenuated viral vaccines, including vaccinia viruses, such as Modified Vaccinia Ankara virus (MV A) for smallpox, and the oral poliovirus vaccine (Sabin), are capable of inducing proper T-cell mediated anti-viral immune responses. These vaccines start by targeting epithelial or fibroblast cells and induce apoptosis, rather than the lytic cell death characteristic of the virulent wild-type strains. These apoptotic cells then recruit monocyte-derived and tissue-resident macrophages through chemoattractant factors such as caspase-dependent secretion of ATP (Elliott et al. (2009) Nature 461:282-286). Virally-infected apoptotic cells then are phagocytosed by macrophages, which sense the presence of viral cytosolic DNA or RNA and induce type I interferon (IFN), which produces an antiviral signaling cascade that recruits and activates CD8 ⁺ T-cell priming (Royo et al. (2014) J Virol 88:5511-5523). Virally-infected macrophages then migrate to the lymph node, where they prime CD4 ⁺ helper T-cells, to promote germinal center B-cell antibody production, and CD8 T-cells, which then traffic to the infected tissue and remain as long-lived tissue-resident memory CD8 ⁺ T-cells. These immune cells are critical for recognizing early infection of tissue-resident cells and eliminating them via FasL- induced (Fas ligand or CD95L or CD178, a type-II transmembrane protein in the TNF family) apoptosis, often before the virus is detected by the immune system (Hobbs et al. (2018) Curr Opin Virol. 28:12-19 El-Jesr etal. (01 October 2020) Front Immunol doi.org/10.3389/immun.2020.568412; Wahid s al. (2005) J Virol 79:401-409).

Macrophages are the primary immune cells that induce interferon-beta (IFNP) and CD8 ⁺ T-cell activation; whereas dendritic cells primarily produce IFNa and activate CD4 ⁺ T-cells (Corrales etal. (2015) Cell Reports 11:1018-1030; Wahid et al. WFf J Virol 79:401-409). The intersection of apoptotic cells, macrophage phagocytosis, and induction of type I IFN to prime CD8 ⁺ T-cells is a hallmark of the ability of a vaccine to generate life-long humoral and cellular immunity. Similarly, defects in cytosolic nucleases (e.g., DNasell, TREX1) can allow leaked or phagocytosed nuclear DNA to trigger cytosolic DNA sensing via cyclic-GA- Synthase (cGAS), which produces a cyclic dinucleotide (cGAMP) that activates Stimulator of IFN Genes (STING). This results in type I IFN production and CD8 ⁺ T-cell priming of either self-antigens to induce autoimmunity, or tumor antigens to induce anti-tumor immunity (Barber et al. (2015) Nat Rev Immunol 15:760-770; Ahn etal. (2018) Cancer Cell 33:862-873).

Designing cancer immunotherapies that convert the immunosuppressive tumor-associated macrophages (TAMs) to Type I IFN-producing macrophages, capable of in situ priming of CD8 ⁺ T-cells to tumor antigens and inducing durable anti-tumor immunity, are needed to address T-cell excluded and immune desert tumors. Immunostimulatory bacteria provided herein address this need. The immunostimulatory bacteria provided herein, as described herein, are genome- modified to remove undesirable bacterial sensing from Toll-like receptor-2 (TLR2), TLR4, and/or TLR5. As described herein, sensing from these TLRs, particularly TLR2, as well TLR4 and TLR5, directly suppress the ability of a macrophage to produce type I IFN and induce pro-inflammatory cytokines which impair CD8 ⁺ T-cell priming. The immunostimulatory bacterium also are modified to include a purine auxotrophy, which provides tumor targeting following systemic, such as IV, dosing. The immunostimulatory bacterium also are genome-modified, such as by modifications resulting in the elimination of flagella, to be taken up only by phagocytic tumor-resident myeloid cells, including macrophages. Among the immunostimulatory bacteria described and provided herein include those that, by virtue of genome modifications, infect phagocytic cells, generally at least to the same extent or in increased amounts compared to the bacteria without the modifications. Included are bacteria that do not infect epithelial cells and/or endothelial cells. The macrophages then destroy these attenuated bacteria, providing plasmid transfer and expression of the payloads as described herein. Of particular interest are combination encoded payloads, such as the combination of a cytokine, such as IL- 15, particularly in the engineered IL-15/IL-15R alpha chain complex (human IL- 15 cytokine fused with the IL-15 receptor alpha chain (IL-15Ra-IL-15sc), and variants of cytosolic DNA/RNA sensors to render them constitutive, such as STING (eSTING) proteins provided herein that are constitutively active, and, also can be modified or selected to limit the production of immunosuppressive NF-kB signaling in favor of antiviral type I IFN signaling. Exemplary of the cytosolic DNA/RNA sensors is the chimeric STING that includes a gain-of-function mutation(s) and a CTT from a STING protein with lower NF-KB signaling activity, such as the eSTING (engineered STING) designated huSTING tazCTT N154S/R284G. As shown herein, the combination of expression of such cytokines and cytosolic DNA/RNA sensors, such as eSTING, in the tumor microenvironment and in the macrophages overcomes the immunosuppressive tumor microenvironment in T-cell desert/ depleted tumors. As demonstrated in the examples, immunostimulatory bacteria provided colonize the tumor microenvironment and deliver payloads to phagocytic APCs, inducing a durable anti-tumor response. This was observed after a single intravenous dose. For example, an exemplified strain YS1646 asd! pagPI ansB/ csgD containing a plasmid encoding huIL-15Ra-IL-15sc + huSTING N154S/R284G tazCTT, leads to IL- 15 secretion and IFN-beta expression, respectively, in cell lines and primary M2 macrophages. The bacterium is selectively internalized by phagocytic APCs in vitro, and tumor-resident APCs in vivo. Primary M2 macrophages polarized toward a costimulatory and phagocytic hybrid M1/M2 phenotype. The bacterium induces immune reprogramming and remodeling of the tumor microenvironment through myeloid and CD8+ T-cell infiltration and activation. Additionally, as exemplified synergistic antitumor activity was observed in vivo with combination therapy with immunotherapy, such as anti-PD- 1 antibody.

- 7 -

RECTIFIED SHEET (RULE 91) ISA/EP Among the immunostimulatory bacteria provided herein are those that target tumor-resident tumor-associated macrophages (TAMs) following systemic, such as IV, administration. Their activity is primarily limited to or is limited to phagocytic and proliferating TAMs, in which they transfer DNA plasmids to the nucleus of a dividing cell, where encoded payloads are expressed. Of interest herein, are the payload combinations, such as combinations of a cytokine, such as an IL-15, particularly IL-15/IL-15R alpha chain complex, and a STING protein, particularly one that has constitutive activity, and also can have lower NF -KB signaling activity compared to wild-type human STING, such as a wild-type human STING of any of SEQ ID NOs. 305-309, which set forth the sequences of allelic variants of human STING proteins with substantially the same activities and properties. Also of interest are immunostimulatory bacteria that encode a type I interferon as a single payload or in combination with one or more other immunostimulatory protein(s), such as the IL- 15/IL-15R alpha chain complex and/or a modified STING, such as a constitutive STING, as described herein.

It is shown herein that the production of the combined payload, such as huIL- 15Ra-IL-15sc and the chimeric STING, such as huSTINGN154S/R284GtazCTT, by the immunostimulatory bacteria, and other alternative immunostimulatory payloads induces a hybrid macrophage phenotype that is particularly advantageous for antitumor therapy. The hybrid macrophage phenotype possesses the immunostimulatory and anti -tumor properties of an Ml or Ml -like macrophage, while retaining the phagocytic properties of an M2 or M2 -like tumor-associated macrophage (TAM). Infected macrophages, as described herein, also have and can be identified by a hybrid SPP1 ⁺ and C1QC ⁺ (expression of both SPP1 and C1QC) macrophage phenotype, with enhanced phagocytic and proliferating properties. The result is a hybrid macrophage phenotype that can phagocytose apoptotic tumor cells, induce constitutive type I IFN to recruit and prime tumor antigen-specific CD8 ⁺ T-cells, and induce durable anti-tumor immunity.

Uptake of the immunostimulatory bacteria modified as described herein to lack flagella and to have modified LPS, as well as other modifications is specific to phagocytic cells, particularly human M2 macrophages. These bacteria comprise genome modifications that remove inflammatory surface components, including flagella, curb fimbriae, and inflammatory LPS. They also include adenosine auxotrophy providing obligatory dependence on nutrients, such as adenosine, ATP, AMP, and purines, that accumulate in the TME, and that are immunosuppressive.

For embodiments in which the encoded payloads are expressed by the eukaryotic host cell transcriptional machinery, expression of the payloads occurs in proliferating macrophage; the plasmids can enter the nucleus in proliferating macrophages. It also is shown and described herein, that the immunostimulatory bacteria provided herein are internalized by M2 macrophages, but not by HUVECs; whereas VNP20009 is internalized by HUVECs and M2 macrophage, indicating that the immunostimulatory bacteria provided herein that lack flagella and have pentaacylated LPS are more specific, and better colonize tissues of interest for anti-tumor treatment, and also as vaccines per se. The bacteria are internalized and trafficked to acidic lysosomes and nuclei in human macrophages.

It is shown herein bacteria and therapeutics described and/or provided herein induce a hybrid M1/M2 phenotype; the resulting hybrid M1/M2 phenotype induces anti-tumor immunity. The M1/M2 phenotype can be identified by markers, indicative of Ml and/or M2 phenotypes, that are upregulated and/or downregulated following treatment. Markers detailed herein, include co- stimulation markers, such as CD80, CD86, and phagocytic markers, such as CD206. Thus, a therapeutic that can deliver a non-integrating (non-integrating into the genome) therapeutic payload to a macrophage and convert or produce or result in macrophage with this phenotype is capable of inducing durable anti-tumor immunity. Durable anti-tumor immunity is evidenced by dose-dependent anti-tumor response upon tumor rechallenge results. The therapeutics, including the immunostimulatory bacteria reverse the immunosuppressive tumor microenvironment. The bacteria also are shown herein to convert immunotherapy, such as anti-PD-1, refractive tumors to responsive tumors. For example, synergistic activity with anti-PD-1 was observed.

It is shown herein, that the presence of proliferating macrophage in a tumor can be prognostic of the effectiveness a therapy that delivers a non-integrating nucleic acid payload. Introduced DNA, such as from the immunostimulatory bacteria, can be transcribed and translated in proliferating macrophage. Proliferating macrophages in a tumor can be identified from a biopsy by prospective biomarkers. Proliferating macrophage can be identified by some or all the following markers:

Tumor gene expression of G2M module (>14 genes of the set), Stathminl (STMN1); Biopsy surface markers: CD68 + KI67 and/or PCNA, MERTK;

SPP1 in some tumor types: lung, gastric; and/or C1QC in some tumor types: colon and breast. Provided herein are methods for identifying proliferating macrophages in human tumors, and for identifying subjects, who will be responsive to treatment with immunostimulatory bacteria, particularly those with a payload that comprises a cytokine and STING is described. Also provided are methods to therapeutically induce an optimal tumor macrophage phenotype prior to treatment by administering a therapy that induces apoptosis.

Immunostimulatory bacteria, therapeutics, and methods

Provided are methods of treating a cancer or tumor administering a therapeutic that, upon administration, results in tumor macrophages that have a hybrid M1/M2 phenotype. The macrophages that that undergo this phenotype change are proliferating macrophages. The therapeutics include any designed or prepared to achieve this phenotypic change. It is shown herein that in embodiments in which the delivery vehicle, such as the immunostimulatory bacteria, encodes an immunostimulatory protein, the nucleic acid must get into the nucleus for transcription. This occurs in proliferating macrophage, such as the immunostimulatory bacteria. The encoded payload, such as combination of immunostimulatory proteins, such as a protein that is part of a cytosolic DNA/RNA sensor pathway that leads to expression of type I interferon (IFN), and also a cytokine, such as, for example, IL-15/IL-15R alpha chain complex, result in the phenotypic change of the macrophage. Proteins that are part of a DNA/RNA sensor pathway include, for example, STING, MDA5, IRF-3, IRF-7, IRF-5, IRF8, and RIG-I, particularly modified forms thereof that result in constitutive expression of type I interferon (IFN). The delivery vehicles also can deliver or encode agonists of these proteins, such as an agonist of one or more of STING, MDA5, IRF-3, IRF-5, IRF-7, IRF-8, and/or RIG-I, to result in type I interferon (IFN) expression.

Also provided are isolated macrophages that have been treated by introducing a therapeutic, such as immunostimulatory bacteria provided herein, that convert the macrophage to an M1/M2 phenotype. Such macrophages can phagocytose apoptotic tumor cells, induce constitutive type I IFN to recruit and prime tumor antigen-specific CD8+ T-cells, and induce durable anti-tumor immunity. Macrophage or immune cells comprising macrophage can be isolated from a subject or they can be, where appropriate, allogeneic, particularly a subject with a tumor or tumors that are immune desert tumors or T-cell excluded tumors, which include suppressive myeloid cells, and for which therapies, such as immunotherapy are ineffective, or a subject treated with but not responding to treatment with immunotherapy, such as a checkpoint inhibitor. The macrophage or cells comprising macrophage can be treated with, such as infected with the therapeutic, such as the immunostimulatory bacteria in vitro, cultured as needed or stored, and then introduced, as in a cell therapy protocol, into a subject to thereby provide macrophage that have anti-tumor activity. The macrophages or cells comprising macrophages that have been treated in vitro to render macrophages M1/M2 hybrids can be administered systemically or intratumorally, or intraperitoneally, or by other suitable route, to result in an antitumor response in the subject. Any of the immunostimulatory bacteria provided herein, that effect this phenotypic conversion, and delivery vehicles, and other therapeutics can be introduced in the cells comprising macrophages. These macrophage can reprogram the tumor microenvironment to result in anti-tumor immunity, including myeloid repolarization. Provided are compositions and cell cultures that comprise macrophage with an M1/M2 phenotype, including compositions formulated for cell therapy. The compositions contain macrophage that, for example, have be infected in vitro with any of the bacteria provided herein that induce a type I interferon (IFN), such as, for example, the bacteria encoding DNA/RNA sensor proteins and/or cytokines, and bacteria that encode a type I interferon (IFN) or a plurality of such interferons.

Exemplary of the therapeutics are the immunostimulatory bacteria described and provided herein. It also is shown herein that the cancers and/or tumors susceptible to treatment with the immunostimulatory bacteria provided herein include those that have one or more of elevated adenosine, TGFbeta, relative to non-tumor tissue, and/or the tumor is hypoxic. Exemplary of such cancers is chronic lymphocytic leukemia (CLL) or a myeloid malignancy. The immunostimulatory bacteria as shown and described herein, colonize solid tumors.

Provided and described herein are immunostimulatory bacteria that are cancer therapeutics by virtue of their ability to effectively colonize tumors, particularly tumor resident immune cells, and by virtue of the encoded payloads that result in an antitumor immune response. As described below, these bacteria, when administered, infect or are taken up by phagocytic cells, such as macrophages in the tumors. As

- 11 -

RECTIFIED SHEET (RULE 91) ISA/EP described herein, bacteria and therapeutics described and provided herein can convert the phenotype of the macrophages to an M1/M2 hybrid phenotype. Because of their ability to colonize tumors and the tumor microenvironment, immunostimulatory bacteria provided herein can be used to treat immune desert (immune-excluded or T- cell excluded or cold) tumors, which have scarce or absent T-cell infiltration in the tumor microenvironment and tumors. These include tumors with stromal barriers. The immunostimulatory bacteria provided herein can turn so-called “cold” tumors, which are resistant to or non-responsive to immunotherapy, into “hot” tumors.

The immunostimulatory bacteria include genome modifications described herein, such that they are TLR2/4/5 attenuated, such as by virtue of elimination of flagella, the msbB'lpagP' phenotype, as well as additional mutations, such as the elimination of curli fimbriae, and also other mutations, such as the ansB' phenotype, described herein. These bacteria can proliferate in vivo. The payloads encoded in the plasmid(s) in the bacteria are those that are part of a cytosolic DNA/RNA sensor pathway that leads to expression of type I interferon (IFN). In particular, these products include one or more mutations, such as gain-of-function mutation(s), to render expression of type I IFN constitutive. These immunostimulatory bacteria also can encode a cytokine or cytokines, such as an IL-15, particularly as an IL-15/IL-15R alpha chain complex, and can encode tumor-associated antigens and/or bi-specific T- cell engager antibodies. They can also encode a type I interferon (IFN) and/or other immunostimulatory proteins. For cancer therapy, the bacteria can be administered systemically.

It is shown herein, that immunostimulatory bacteria, such as the bacteria that have genome modifications as described herein that target or accumulate in macrophage, such as the bacteria designated STACT (S. Typhimurium-Attenuated Cancer Therapy). The STACT contain particular genome modifications, such as no flagella, modified LPS, and other properties. The STACT bacteria contain a plasmid. The plasmid encodes a payload of interest, such as one or more immunostimulatory proteins, such as a cytokine and cytosolic DNA/RNA sensor, such as STING, generally modified to have constitutive activity so that, upon infection of macrophage and expression of the encoded STING, type I interferon is constitutively expressed. For example, the bacteria include a plasmid that encodes a cytokine, such as an 15/IL- 15R alpha chain complex and a modified STING that constitutively induces type I IFN. These bacteria specifically target and reprogram immunosuppressive tumor- associated macrophages (TAMs) into hybrid M1/M2 phenotype phagocytic and T-cell priming macrophages.

Included are bacteria that, as described below, encode the payload(s) under control of a prokaryotic promoter. In some embodiments, the nucleic acid includes or encodes signals that prevent or do not facilitate translation in the bacteria, whereby the RNA is delivered to the tumor microenvironment and tumors and immune cells, including the macrophages, therein. In other embodiments, the RNA is transcribed by the bacterial ribosomes and proteins are delivered.

The immunostimulatory bacteria provided herein, when administered to a subject, convert immune-suppressive tumor-associated macrophages (TAMs) into tumor antigen presenting cells (APCs), capable of inducing type I interferon (IFN)- mediated recruitment and in situ priming of CD8 ⁺ T-cells. The immunostimulatory bacteria, when administered to a subject with cancer, effect the phenotype conversion and induce durable anti-tumor immunity in T-cell excluded and immune desert solid tumors. The immunostimulatory bacteria, such as the bacteria that are modified to target macrophages and that encode a combination of a protein that constitutively induces type I interferon expression in macrophages, and cytokine, can effect phenotypic changes. These changes can convert T-cell excluded/desert tumors into hot tumors, which are susceptible to treatment with immunotherapy, such as anticheckpoint antibodies. The immunostimulatory bacteria provided herein that encode one or more type I interferon(s) (IFN(s)) also can be used to convert T-cell excluded and desert tumors into hot tumors that are responsive to immunotherapy, such as by therapy with immune checkpoint inhibitors.

The immunostimulatory bacteria herein have genome modifications that render them auxotrophic for purines and purine metabolites, particularly adenosine. Purines and metabolites accumulate to pathologic concentrations in tumors, and not in healthy tissue; among the purine metabolites is adenosine, which is immunosuppressive. The immunostimulatory bacteria, which accumulate in the tumor microenvironment and tumors, thus reduce the concentrations and reverse or prevent the immunosuppressive effects of accumulation of metabolites, such as adenosine.

Immunostimulatory bacterium bacteria provided herein are genome-modified to have reduced bacterial component recognition by TLR2, TLR4 and TLR5; these modifications reduce or prevent production of pro-inflammatory cytokines that suppress CD8 ⁺ T-cell priming, as well as TLR-mediated signaling pathways that

- 13 -

RECTIFIED SHEET (RULE 91) ISA/EP impair macrophage induction of type I IFN. Induction of type IFN can be rendered constitutive by the expression of the bacterially-encoded modified cytosolic DNA/RNA sensors, such as STING, particularly modified STING that renders type I interferon expression constitutive in the macrophage. Following phagocytosis of administered bacteria, such as administered by intravenous administration, the bacteria are rapidly eliminated and the plasmid-encoded immunomodulatory payloads are ectopically expressed. The payload delivery is limited to immunosuppressive TAMs of the tumor microenvironment, and not to phagocytic macrophages in the liver or other tissues, due to, as shown herein, the requirement for bacterial uptake and DNA plasmid transfer to the nucleus that the macrophage are phagocytic and proliferating

The immunostimulatory bacteria provided herein and the methods and uses herein address an unmet need for treating tumors that are macrophage-rich but lack T- cells and do not respond to existing immunotherapies. The bacteria provided herein are taken up by phagocytic cells and effect the conversion to M1/M2 hybrid phenotype. It is shown herein that proliferating macrophages can transfer the bacterial plasmid into the nucleus, and transcribe the encoded payload, which are then translated to produce immunostimulatory or immunomodulatory proteins, such as the modified STING and a cytokine. Exemplary of these bacteria are referred to as STACT. The STACT contain a plasmid-encoded human IL-15 cytokine fused with the IL- 15 receptor alpha chain (IL-15plex) and an engineered constitutive STING (eSTING). The working examples herein demonstrate that such bacteria promote CD8 ⁺ T-cell mediated tumor clearance in T-cell excluded tumors and elicit durable anti-tumor immunity, and also have a highly favorable safety profile following IV dosing in primates. The STACT-delivered combination of IL-15plex + eSTING to immunosuppressive TAMs induces a heretofore unknown hybrid M1/M2 macrophage phenotype. Macrophages with this phenotype exhibit the immunostimulatory and T- cell priming properties of an Ml -like macrophage, and retain the tumor cell phagocytic properties of an M2-like TAM. As shown herein, the infected macrophages have a hybrid SPP1 ⁺ C1QC ⁺ phenotype, with enhanced phagocytic and proliferating properties. The resulting macrophages with this phenotype phagocytose apoptotic tumor cells, induce type I IFN, produce IL-15 to recruit, prime and maintain tumor antigen-specific CD8 ⁺ T-cells, and promote durable anti-tumor immunity. As described above, in exemplary embodiments, among the immunostimulatory bacteria provided herein are those that are referred to with the acronym STACT (S. Typhimurium-Attenuated Cancer Therapy). These are exemplary of the immunostimulatory bacteria described and provided herein. The immunostimulatory bacteria, including the exemplary bacteria designated STACT, comprise genome modifications that result in advantageous properties, including, but not limited to: (1) enhanced, compared to the unmodified parental strain VNP20009 (also designated YS1456), tolerability after IV dosing, (2) tumor-specific enrichment, (3) phagocytosis by tumor-resident antigen-presenting cells (APCs) with a lack of epithelial cell infectivity, (4) multiplexed genetic cargo delivery, and (5) attenuation of bacterial pathways that impair CD8 ⁺ T-cell function. As a particular example of the exemplary STACT immunostimulatory bacteria are those that encode immunomodulatory molecules, including immunostimulatory proteins, such as a cytokine, such as an IL-15, such as IL-15/IL-15R alpha chain complex, and a modified STING protein that constitutively induces type I IFN. Exemplary of STACT immunostimulatory bacterium are the strains that are designated YS 1646Aavc//AFLG/A/?agE/A««s5/Ac.sg _J D.

It is shown herein, that the presence of proliferating macrophage in a tumor can be prognostic of the effectiveness of a therapy that delivers a non-integrating nucleic acid payload. Proliferating macrophage in a tumor can be identified from a biopsy by prospective biomarkers. Proliferating macrophage can be identified by the following markers:

Tumor gene expression of G2M module (>14 genes of the set), Stathminl (STMN1);

Biopsy surface markers: CD68 + KI67 and/or PCNA, MERTK;

SPP1 in some tumor types: lung, gastric;

C1QC in some tumor types: colon and breast.

Provided herein are methods for identifying proliferating macrophages in human tumors, and for identifying subjects, who will be responsive to treatment with immunostimulatory bacteria, such as with the bacteria encoding a payload that, when expressed, comprises a cytokine and STING is described. Also provided are methods to therapeutically induce an optimal tumor macrophage phenotype prior to treatment by administering a therapy that induces apoptosis.

- 15 -

RECTIFIED SHEET (RULE 91 ) ISA/EP Immunostimulatory bacteria encoding other combinations of encoded immunostimulatory proteins are provided. These include cytokines, type I interferon (IFN)-inducing factors, co- stimulatory receptors, checkpoint antibodies and TGF0R- Fc decoys. The engineered STING (eSTING) proteins as described herein can have constitutive type I IFN inducing activity, and also can have low NF-KB signaling activity, relative to wild-type human STING. Combinations of the eSTING and immunostimulatory protein were evaluated in primary human APCs using in vitro functional assays. Numerous combinations possessed the desired activities. Among these are the bacteria that encoded IL-15Ra-IL-15 (IL-15) + eSTING, which were evaluated in murine tumor models for therapeutic efficacy and mechanism, as well as tolerability in rodents and primates after systemic administration. Treatment with bacteria encoding the combination of modified STING (eSTING) protein and cytokine exhibited a high degree of complete tumor responses that were entirely CD8 ⁺ T-cell dependent. In an autochthonous breast cancer model that lacks any significant lymphocyte infiltrate, these bacteria uniformly enrich in each spontaneous lesion to high levels after IV dosing and resulted in significant CD8 ⁺ T-cell infiltration. In primates, these bacteria were well-tolerated, rapidly cleared, and elicited minimal cytokine response after IV dosing. Thus, the bacteria, so-engineered, deliver a payload of the combination of a cytokine, such as an IL- 15 (IL-15/IL-15R alpha chain complex), and eSTING to phagocytic APCs in the solid tumor microenvironment after systemic administration. The bacteria promote CD8 ⁺ T-cell mediated tumor clearance in T-cell excluded tumors, elicit durable anti-tumor immunity, and are well -tolerated in primates.

Tumors that are particularly susceptible to treatment are shown herein that are enriched in adenosine (AD) or ADO pathway metabolites (ATP, AMP, Adenine, guanine, and adenosine) myeloid signatures. These include for example tumors that have one or more of elevated adenosine and TGFbeta, relative to a non-tumor tissue, and/or hypoxic tumors or cancers. These include lymphocytic leukemia and myeloid malignancies. They also include tumors that have these signatures, including renal clear cell carcinoma, mesothelioma, breast, pancreatic, NSCLC adenocarcinoma, sarcoma, ovarian, cervical, endocervical, head and neck squamous, esophageal adenocarcinomas, stomach carcinoma, NSCLC squamous tumors, and some thyroid tumors. These tumors can be treated with the therapeutics provided herein to render them hot tumors.

Provided are immunostimulatory bacteria that contain genome modifications and a plasmid that encodes one or more therapeutic products, such as anti-cancer therapeutics or associated treatments. The genome modifications result in immunostimulatory bacteria that accumulate in the tumor microenvironment and in tumor-resident immune cells, where they express the encoded therapeutic products. The immunostimulatory bacteria provided herein encode one or a plurality of complementary products that stimulate or induce or result in a robust anti-cancer response in the subject. As demonstrated in the examples, the immunostimulatory bacteria provided and described herein reprogram the immunosuppressive tumor microenvironment to an anti-tumor phenotype leading to T-cell infiltration and activation, B-cell infiltration, macrophage repolarization and activation, dendritic cell activity, which induce potent antigen-specific CD8+ T-cell responses. The immunostimulating payload(s), such as the engineered STING and cytokines, results in expression and production of cytokines and other factors leading to anti-tumor immunity, including MHC upregulation.

Provided are methods of treating a tumor, comprising administering a therapeutic that, upon administration, results in tumor macrophages that have a hybrid M1/M2 phenotype. Exemplary of the methods are those where the resulting M1/M2 macrophages are capable of phagocytosing an apoptotic tumor cell and/or a delivery vehicles. Provided are methods of treating a tumor with a therapeutic, by identifying a subject whose tumor comprises proliferating macrophages; and administering the therapeutic that delivers a payload into the proliferating macrophages and converts them into macrophages with an M1/M2 hybrid phenotype. Provided are therapeutics for use for treatment of tumors in a subject, where: the therapeutic converts proliferating macrophages into M1/M2 hybrid phenotype macrophages; a tumor in the subject had been identified as comprising proliferating macrophages, and the therapeutic comprises a delivery vehicle that has attenuated TLR2, TLR4, and/or TLR5 activity, whereby production of type I IFN by macrophages that comprise the therapeutic in not inhibited. It is shown herein that TLR2, TLR4, and TLR5 activity or response inhibits expression of type I IFN, such as in macrophages. Hence the therapeutics have attenuated TLR2 or TLR2, TLR4 and TLR5 activity.

- 17 -

RECTIFIED SHEET (RULE 91) ISA/EP Provided are therapeutics that are effective for converting macrophages into an M1/M2 hybrid macrophage. The therapeutics comprise: a delivery vehicle that has attenuated TLR2, TLR4, and/or TLR5 activity, whereby production of type I IFN by macrophages that comprise the therapeutic in not inhibited; and nucleic acid encoding at least two different immunostimulatory proteins, wherein one protein induces type I IFN when introduced into macrophages, and the other stimulates anti-viral or anticancer immune responses. The nucleic acid generally is provided in a form, such as in a non-integrative plasmid, that does not integrate into the host cell genome, such as by integration into a chromosome in the genome.

Provided are methods of converting an immune excluded or immune desert tumor into a T-cell infiltrated tumor, comprising administering a therapeutic, such as those described above and herein, including the immunostimulatory bacteria that have attenuated TLR2/4/5 activity and encode an immunostimulatory protein, into the tumor. The therapeutics, thus, can be used for converting an immune desert tumor into a T-cell infiltrated tumor. Encoded immunostimulatory proteins include, for example, a cytokine, such as IL-15/IL-15R alpha chain complex, and/or a type I interferon (IFN), such as interferon-alpha or interferon-beta, and combinations of the cytokines.

Provided are methods for detecting subjects likely to or predicted to respond to treatment with a therapeutic comprising a delivery vehicle and non-integrating nucleic acid encoding one or more immunostimulatory proteins, comprising detecting proliferating macrophages, and/or detecting particular markers in a tumor or body fluid sample. Thus, the methods identify a subset of subjects in which the therapeutic is likely to be effective, and excluding subjects in whom it is not likely to be effective. This method can comprise CD68 and PCNA, and/or Ki67 to identify a subject predicted to or likely to respond to treatment with a therapeutic comprising a delivery vehicle and non-integrating nucleic acid encoding one or immunostimulatory proteins.

These methods, therapeutics, and uses, as well as methods for identifying subjects likely to be responsive to treatment with a therapeutic that converts a macrophage into the M1/M2 hybrid phenotype, include those where response to treatment and/or proliferating macrophage are identified by a combination of markers detectable by immunohistochemistry (H4C) and genetic markers for a particular tumor type. These methods can be effected by obtaining a tumor biopsy or body fluid sample, and detecting proliferating macrophages in the biopsy or sample or detecting a combination of markers detectable by IHC and genetic markers for a tumor type in a subject with a particular cancer or tumor. For example the markers that can be detected by IHC markers ClQC ⁺ or SPP1 ⁺ . These markers, shown in the examples, are correlated with particular cancer types, and their use as prognostic markers varies by cancers are shown herein. It also is shown herein that C1QC ⁺ or SPP1 ⁺ , depending on tumor or cancer type, can be combined with genetic markers, to identify subjects whose tumors are likely to respond to the therapeutics descried and provided herein. As described herein, these therapeutics are identified as therapeutics that convert macrophages into M1/M2 hybrid phenotype tumors, which are responsive to these therapeutics. Hence the markers and methods described herein for identifying proliferating macrophage and/or tumors with particular markers, identify those in which macrophages will be converted to the hybrid M1/M2 phenotype. Tumors that comprise such macrophage or that have the markers are susceptible to treatment with the therapeutics described and provided herein.

In some embodiments of the methods, therapeutics and uses are those in which combinations of immunohistochemistry markers and genetic markers for tumor types are used to select subjects for treatment, who are then treated with the therapeutic herein that result in macrophages with the M1/M2 hybrid phenotype. Exemplary combinations of makers and tumor types include, but are not limited to:

SPP1 ⁺ and NRF2 pathway alterations in a tumor biopsy or body fluid sample from a subject with a squamous carcinomas, such as, for example a squamous carcinoma selected from among selected lung (LUSC), head and neck (HNSC), cervical (CESC), esophageal (ESCA), bladder (BLCA), and kidney renal papillary (KIRP);

SPP1 ⁺ and TP53 mutations in breast cancer (BRCA);

SPP1 ⁺ and PI3K mutations in prostate cancer (PRAD);

SPP1 ⁺ and BRAF mutations in skin cutaneous melanoma (SKCM);

C1QC ⁺ and HIPPO pathway mutations;

C1QC ⁺ in uterine corpus endometrial cancer (UCEC);

C1QC ⁺ and KMT2A mutations in bladder cancer (BLCA); and C1QC ⁺ and TP53 pathway mutations in breast cancer (BRCA). The therapeutics that are provided and used in the methods and uses are therapeutics that are a tumor-targeted therapy requiring or mediating nucleic acid transfer to immune cells for non-integrating ectopic gene expression; and tumor- targeted therapy is a therapy that is directed to or that accumulates in or is taken up by tumors, the tumor microenvironment, and or tumor-resident immune cells. For example the therapeutics comprise a delivery vehicle and nucleic acid encoding a immunostimulatory protein; the therapeutic has attenuated TLR2 activity, whereby type I IFN is not inhibited in macrophages that comprise the therapeutic or encoded nucleic acid. For example, provided are therapeutics, methods, and uses where: the therapeutic comprises a delivery vehicle and nucleic acid encoding a immunostimulatory protein; and the therapeutic has attenuated TLR2 and TLR4 or TLR2/4/5 activity, whereby type I IFN is not inhibited in macrophages that comprise the therapeutic or encoded nucleic acid.

Provided are methods for identifying therapeutics that convert macrophages to an M1/M2 phenotype, comprising: a) preparing one or more candidate therapeutics that comprise a delivery vehicle and nucleic acid encoding immunostimulatory proteins, wherein one of the immunostimulatory proteins is induces an anti-viral or anti-cancer immune response, and the other induces type I IFN, and the delivery vehicle is TLR2 or TLR4 or TLR2 and TLR4 or 5, or TLR2/4/5 attenuated, whereby the therapeutic does not inhibit type I IFN in macrophages when introduced into or that infect the macrophages; b) introducing the candidate therapeutic(s) into proliferating macrophages; c) determining the phenotype of the resulting macrophages; and d) selecting a candidate therapeutic(s) therapeutic if the resulting macrophage have a M1/M2 hybrid phenotype.

Provided are methods of treatment of cancer, comprising administering to a subject, identified as having proliferating macrophage and/or the markers and genetic markers described herein as prognostic of effectiveness of the therapeutics provided herein, a therapeutic that was identified by the above method. Exemplary of methods, therapeutics, and uses provided herein are those where: the therapeutic comprises a delivery vehicle and nucleic acid encoding at least two immunostimulatory proteins: one of the immunostimulatory proteins induces or results in expression of type I IFN in the proliferating macrophages; and another of the immunostimulatory proteins induces or results in expression of anti-cancer or anti-viral cytokines or chemokines or other anti-cancer or anti-viral immunostimulatory effectors. Other exemplary methods, therapeutics and uses are those where: the therapeutic comprises a delivery vehicle containing nucleic acid encoding an immunostimulatory protein that constitutively induces type I IFN in the macrophages; the vehicle does not induce or has reduced TLR2 or reduced TLR2/4/5 induction/response such that type I IFN is not inhibited; and the nucleic acid encoding the unmunostimulatory protein is transcribed and translated in the macrophages. Other examples are those where: the therapeutic comprises a delivery vehicle; and the delivery vehicle is a bacterium, a nanoparticle, a virus, or an exosome. For example, the therapeutic can comprise a delivery vehicle selected from among a nanoparticle, a virus, an exosome, a cell, and a bacterium, optionally with the proviso that the delivery vehicle is not a bacterium, or is not a Salmonella species, or is not a STACT species. In other examples, the therapeutic comprises a delivery vehicle and nucleic acid; the delivery vehicle is a lipid nanoparticle, or an attenuated bacterium, or an immune cell, or an oncolytic virus; and the delivery vehicle does not or is modified to attenuate TLR2 activity, whereby expression of type I IFN in the macrophage is not inhibited comprising the therapeutic or delivery vehicle.

Provided are methods, therapeutics, and uses where the M1/M2 phenotype markers comprise: a) at least two of any of the following markers: Hybrid Markers (lower than M2, higher than Ml): SPP1, CD209, CD206; and Induced Markers: MERTK, C1QC, IFN-a2a, IFNpl, CXCL10, 4-1BBL (TNFSF9), MYC; and/or b) wherein uptake of the therapeutic by M2 macrophage induces a hybrid M1/M2 phenotype that retains M2 phagocytic capacity, upregulates Ml-like costimulatory receptors (CD80/86) and lymph node chemotaxis receptors (CCR7), and produces type I IFN-mediated cytokines and chemokines. For example, the macrophage M1/M2 hybrid phenotype markers comprise CD209 and CD206 at levels lower in the resulting macrophage than in M2 macrophage and higher than in Ml macrophage. Other combinations of markers for identification of macrophage with the M1/M2 hybrid phenotype are those where phenotypic markers comprise all of:

Hybrid Markers (lower than M2, higher than Ml): SPP1, CD209, CD206 and/or two or more Induced Markers: MERTK, C1QC, IFN-a2a, IFNpl, CXCL10, 4- IBBL (TNFSF9), MYC; or those where the phenotypic markers comprise:

Hybrid Markers (lower than M2, higher than Ml): SPP1, CD209, CD206 and/or all of Induced Markers: MERTK, C1QC, IFN-a2a, IFN l, CXCL10, 4-1BBL (TNFSF9), and MYC. For example, the macrophage phenotype that is induced or results from treatment with the therapeutics herein, such as a therapeutic that has attenuated TLR2/4/5 activation so that type I IFN expression is not inhibited, and that encodes in a non-integrating nucleic acid vehicle, such as a plasmid, IL-15/IL-15R alpha chain complex + eSTING (a STING described herein that has constitutive activity), the marker profile post-treatment is: or, wherein markers post-treatment that are upregulated in the resulting macrophages are co-stimulatory molecules CD80, CD86, chemokine signaling CCR7, CXCL10, CXCL11, PRRs (pattern recognition receptors), which are upregulated relative to Ml, downregulated relative to M2 macrophage, CD206, CD209; and scavenger receptors upregulated CD68, CD163.

In the methods, therapeutics, and uses provided herein, proliferating macrophages, such as proliferating M2 macrophages, can be identified by the presence of biopsy surface markers: CD68 + KI67 and/or PCNA, MERTK, and or by gene expression of the G2M module, where half or more than half (> or >14 genes of the set) are expressed, and optionally STMN1 is expressed. The proliferating macrophages also can be identified by STMN 1 + the G2M module with at least half of the genes, such as > or >14 genes of the set, or by the G2M module with at least half of the genes, or by the markers CD68, MERTK, and KI 67 and/or PCNA. For example, the M1/M2 hybrid phenotype is characterized by or identify by Hybrid Markers (lower than M2, higher than Ml): SPP1, CD209, CD206, such as, for example, where the phenotype comprises induced markers: MERTK, C1QC, IFN- a2a, IFNp, CXCL10, 4-1BBL, and/or MYC.

RECTIFIED SHEET (RULE 91 ) ISA/EP In some embodiments, the hybrid M1/M2 macrophage phenotype is characterized by the following markers: Hybrid Markers (lower than M2, higher than Ml): SPP1, CD209, CD206 and/or Induced Markers: MERTK, C1QC, IFN-a2a, IFNP, CXCL10, 4-1BBL, MYC. In others the resulting macrophages are CIQC ^h 'SPP l ^lo " . In all embodiments, the macrophages that comprise the therapeutic or delivery vehicle can be proliferating macrophages, such as proliferating M2 macrophages that, upon expression of the encoded payload in the therapeutic, are converted to M1/M2 hybrid phenotype macrophages.

As described above and herein, the therapeutics that result in the M1/M2 hybrid phenotype in macrophages include those that comprise nucleic acid, generally in a form that is non-integrative, whereby it does not integrate into a host genome. The nucleic acid encodes immunostimulatory proteins, including those and combinations described in this disclosure, such as, but are not limited to, a cytokine and a constitutive STING.

Provided are method of treating a tumor, comprising: identifying a subject whose tumor comprises proliferating macrophages; and administering a therapeutic that delivers a non-integrating genetic payload into the proliferating macrophages, whereby the encoded payload is transcribed.

Also provided are methods of increasing the therapeutic effect of an immunostimulatory bacterium, such as those provided herein, the method comprising administering an apoptosis-promoting agent prior to administration of the immunostimulatory bacterium to a subject. Exemplary of an agent that promotes tumor apoptosis are chemotherapeutic agents, such as, for example, agents selected from among docetaxel (DTX), paclitaxel (PTX), doxorubicin (DOX), 5-fluorouracil (5-FU), carboplatin (CARB), cyclophosphamide (CTX), and other such chemotherap euti cs .

Also provided are methods of increasing the therapeutic effect(s) of an immunostimulatory bacterium in a subject by pre-treatment of the subject with anti- PD-1 treatment to suppress PD-1 expression on macrophages to thereby promote their phagocytic capacity prior to or with the immunostimulatory bacteria, and, after administering the immunostimulatory bacterium, treating with anti-PD-Ll after a sufficient time so that the nucleic acid encoding the payload(s) is delivered to the macrophages, so that PD-L1 is then induced on macrophages. Provided are methods of treatment of cancer and uses of the therapeutics for treatment of cancer. Provided are methods of treatment of cancer, comprising administering an immunostimulatory bacterium to a subject who has been pretreated with an apoptosis-promoting agent prior to the administration of the immunostimulatory bacterium to the subject. Provided are methods of treatment of cancer in a subject, comprising: first treating the subject with an apoptosis-promoting agent; and then administering an immunostimulatory bacterium. These include methods and uses in which the therapeutic, such as the immunostimulatory bacterium, when administered, converts macrophage to an M1/M2 hybrid phenotype. Exemplary of the immunostimulatory bacterium are those that have TLR2, TLR4, and/or TLR5 activity, whereby production of type I IFN by macrophages that comprise the therapeutic is not inhibited; and nucleic acid encoding at least two different immunostimulatory proteins, wherein one protein induces type I IFN when introduced into macrophages, and the other stimulates anti-viral or anti-cancer immune responses.

Also provided are methods and uses of increasing the therapeutic effect of an immunostimulatory bacterium in a subject, comprising: pre-treating the subject with anti-PD-1 antibody or other PD-1 antagonist to suppress PD-1 expression on macrophages in the tumor of the subject to thereby promote their phagocytic capacity; administering the immunostimulatory bacterium, wherein the bacterium encodes one or more immunostimulatory protein(s); and then, after a sufficient time so that the nucleic acid encoding the payload(s) is delivered to the macrophages, treating with an anti-PD-Ll agent. Provided are methods and uses of increasing the therapeutic effect of an immunostimulatory bacterium in a subject, comprising pre-treatment of the subject with anti-PD-1 antibody or other antagonist to suppress PD-1 expression on macrophages to thereby promote their phagocytic capacity prior to or with the immunostimulatory bacteria, and them administering the immunostimulatory bacterium. In these embodiments, the anti-PD-1 treatment can be administered at least about 3, 6, 12, 24, 36, 48, or more hours, such as about two days, before administration of the immunostimulatory bacterium. Anti-PD-1 agents include antagonists and antibodies, which include antibodies and single chain or other forms thereof that bind or inhibit PD-1.

In accord with the method, therapeutics and uses provided herein, subjects that are selected for treatment with the therapeutic that encodes a non-integrating nucleic acid are identified by obtaining a biopsy of a tumor or body fluid from the subject; and selecting a subject for treatment if the phagocytes in the biopsy of the tumor or body fluid are proliferating Body fluids include, but are not limited to, urine, blood, plasma, sweat, CSF, and other such samples. Proliferating macrophage can be identified by detecting the following markers: tumor gene expression of G2M module (>14 genes of the set) alone or +Stathminl (STMN1); and/or biopsy surface markers: CD68 + KI67 and/or PCNA, MERTK; and/or SPP1 in lung or gastric tumors; and/or C1QC in colon and breast cancers. In embodiments of the methods, therapeutics, and uses, the macrophages are M2 macrophages.

The therapeutics herein that convert macrophages into M1/M2 hybrid phenotype macrophages, also can be used to treat fibrotic diseases. The phenotype is effective for treatment of fibrotic disease; hence any of the therapeutics provided herein that convert the phenotype can be used for such treatment.

In the methods, therapeutics, and uses, the therapeutics comprise nucleic acid encoding immunostimulatory proteins; the encoded proteins can comprise a cytokine and a cytosolic DNA/RNA sensor that induces expression of type I IFN; and the cytosolic DNA/RNA sensor is modified to have increased or constitutive activity in inducing type I IFN. The cytosolic DNA/RNA sensor can be modified to have constitutive activity, whereby type I IFN is induced in the absence of ligands and/or cytosolic DNA/RNA. Exemplary cytosolic DNA/RNA sensors and modified forms thereof are described and exemplified below.

The therapeutics, methods and uses include those where therapeutic comprises a delivery vehicle and nucleic acid, such as DNA, where the delivery vehicle has attenuated or eliminated TLR 2, particularly TLR/2/4/5 induction, and encodes a cytokine and a STING pathway protein that constitutively induces type I IFN to result in a hybrid M1/M2 proliferating and phagocytic macrophage phenotype.

In other embodiments, the therapeutics for use in the methods and uses and identified by screening methods are those where: the nucleic acid in the therapeutic encodes a immunostimulatory protein that is selected from among STING, MDA5, IRF-3, IRF-7, and RIG-I; and the immunostimulatory protein comprises a modification(s) that is a gain-of-function (GOF) mutation(s) that renders the STING, MDA5, IRF-3, IRF-7, or RIG-I constitutively active, whereby expression of type I IFN is constitutive. The therapeutics for use herein and in the described and claims methods and uses, include any described herein, or any that comprise nucleic acid, generally nonintegrating, such as in a non-integrating (into the genome) plasmid, encoding an immunostimulatory protein, particularly provided in a delivery vehicle that does not inhibit TLR2, or TLR2, TLR4, and/or TLR5. The nucleic acid can encode any of the payloads and combinations thereof described herein, and the therapeutic includes the immunostimulatory bacteria provided herein or immunostimulatory bacteria known in the art that have the requisite properties.

Because of the similarity in the immune response between an anti-tumor response and an anti-viral response, immunostimulatory bacteria provided herein also can be used to treat infectious diseases. The immunostimulatory bacteria can encode an anti-viral or anti-bacterial therapeutic, such as an inhibitor of a viral or bacterial product, or an inhibitor of the expression of a viral or bacterial product, or a viral or bacterial antigen. The combination of the immune response from the immunostimulatory bacteria and the therapeutic anti-pathogen product, and also to the immunostimulatory proteins and other such therapeutics, provides a therapeutic immunostimulatory bacterium for vaccinating against and/or for treating infectious diseases, particularly diseases associated with viral infections, such as chronic viral infections and latent viral infections. Of interest are chronic viral infections, such as infections by hepatitis viruses, herpesviruses, varicella zoster virus (VZV), Epstein- Barr virus (EBV), human immunodeficiency virus (HIV), human T-cell leukemia virus (HTLV), Respiratory Syncytial Virus (RSV), measles virus, and other such viruses that chronically infect subjects. The immunostimulatory bacteria also can be used for treatment of acute infections as well, such as initial infections with chronic influenza, P. gingivalis, and coronaviruses, such as Severe Acute Respiratory Syndrome coronavirus (SARS-CoV), Middle East Respiratory Syndrome coronavirus (MERS-CoV), and Severe Acute Respiratory Syndrome coronavirus 2 (SARS-CoV-2, which causes COVID-19). Targeted pathogenic bacteria also include, for example, species of Escherichia, Staphylococcus, Pseudomonas, and Porphyromonas .

Also provided herein are immunostimulatory bacteria that can be used and/or formulated as vaccines for administration into tissues, such as by intramuscular injection, inhalation, and other such direct routes. These bacteria are designed to be non-replicating in vivo, and, thus, comprise a nutritional auxotrophy, such as a thyA' so that they do not express active thymidylate synthase, and they encode the payload under control of a promoter recognized in the bacterium. If they are intended to deliver protein payloads to a vaccinated host, the encoded payloads include sequences or are designed so that they are translated in the bacterial host. If they are intended to deliver RNA, then the encoding nucleic acids are designed so that the bacterial ribosomes cannot translate them, but so that eukaryotic ribosomes can translate them. This can be effected, for example, by including an IRES in the encoding nucleic acid. The payloads of the vaccines include nucleic acid encoding the immunizing antigen or protein, such as an antigen from a viral or bacterial pathogen. Payloads also can include immunostimulatory proteins, such as a product, such as STING, particularly modified STING, that is part of a cytosolic DNA/RNA sensor pathway that leads to expression of type I interferon (IFN), and also, optionally, a cytokine, such as an IL- 15, such as IL-15/IL-15R alpha chain complex. The vaccines are formulated for a suitable route of administration, and include aerosols and emulsions, tablets, and powders.

The immunostimulatory bacteria provided herein can encode an antigen or antigens from a pathogen, such as a viral antigen, and are used as a vaccine to prevent infection, or to treat existing infections. Antigens include, but are not limited to, any that are known to those of skill in the art to elicit an immmunoprotective response or to ameliorate a disease resulting from the pathogen. These immunostimulatory bacteria, by virtue of the ability to accumulate in immune cells, such as antigen- presenting cells, can prime the T-cell response to a pathogen, such as a virus. For example, as described in detail herein, among the immunostimulatory bacteria provided herein are those that are deficient in asparaginase II, an enzyme that suppresses the function of T-cells. Any of the immunostimulatory bacteria described and provided herein can be used. For example, it is described and shown herein, that eliminating asparaginase II activity, such as by modifying the bacterial genome to eliminate expression of active enzyme, can be used to encode an antigen or combination of antigens. The resulting bacteria promote an anti-pathogen, such as anti-viral, T-cell response. The combination of expression of an antigen, such as from a pathogen, bacterial or viral or other, with the ability to accumulate in immune cells, such as antigen-presenting cells, provides protection from infection by the pathogen. For example, the immunostimulatory bacteria can encode a viral antigen, such as an antigen from an essential viral core protein shared among a family of viruses or across viral families. For example, in the case of a coronavirus, such as SARS-COV2, an

- 27 -

RECTIFIED SHEET (RULE 91) ISA/EP antigen from the nucleocapsid and/or non-structural M proteins can enhance CD8 T- cell responses to heavily conserved and less mutated core proteins, thereby providing broad pan-coronavirus protection, to provide effective vaccines and treatments. Proteins and antigens from this corona virus and corona virus family that are used for immunization and/or treatment are known, and exemplary ones are described herein and known to those of skill in the art. In addition to spike proteins, portions thereof, and modified spike proteins, other proteins have been identified for this purpose. See, e.g., Cohen et al., (2021) Cell Reports Medicine 2:1000354.

The immunostimulatory bacteria can encode an anti-viral therapeutic or antibacterial therapeutic. Such therapeutics include inhibitors of viral genes and proteins, such as proteins required for replication and/or packaging, or the immunostimulatory bacteria can encode a therapeutic that prevents binding or interaction of a virus with a receptor or receptors that facilitate or provide for viral entry into a target cell. In some embodiments, expression of the encoded therapeutic protein, such as the antigen or antigenic protein, can be under the control of a prokaryotic promoter. In other embodiments, the protein can be expressed under control of a eukaryotic promoter. The choice of promoter depends upon whether it is to be expressed in the bacterium, such as before administration as described herein for delivery of mRNA that is translated in the host, or the protein is to be expressed in the host cells, such as immune cells, after delivery.

The immunostimulatory bacteria provided herein include genome modifications, such as deletions, disruptions, and other alterations that result in inactive encoded product, such as changing the orientation of all or part of the gene, so that functional gene products are not expressed. Among the immunostimulatory bacteria provided are those that are modified so that the resulting bacteria are msbB' /purl'. In some embodiments, the bacteria are msbB' and purl', whereby the full length of at least the coding portion of the msbB and/or purl genes are/is deleted. The genome of the bacteria also can be modified so that the bacteria lack flagella. This is effected in bacteria that normally express flagella. In such bacteria, for example, the fliC and fljB genes in Salmonella, or equivalent genes in other species to fliC and fljB, can be deleted or otherwise modified so that functional gene product is not expressed. The bacteria also can be modified so that they are adenosine auxotrophs, and/or are msbB' IpagP' . Also provided are immunostimulatory bacteria and pharmaceutical compositions containing them, where the bacteria do not express L-asparaginase II, whereby the bacteria are ansB'. Elimination of the encoded asparaginase activity improves or retains T-cell viability /activity. Therapeutic bacteria, such as inactivated or attenuated bacteria that are used as vaccines, can be improved by modifying the bacterial genome to eliminate asparaginase activity. Exemplary of such vaccines is the BCG (Bacillus Calmette-Guerin) vaccine and related vaccines, which are used to immunize against tuberculosis. The BCG vaccine is known to have variable effectiveness; eliminating the asparaginase can improve the effectiveness of such vaccine, because the endogenous bacterial asparaginase inhibits or reduces T-cell activity.

Provided herein are immunostimulatory bacteria that contain a plasmid encoding a therapeutic product, or combinations of therapeutic products, under control of a eukaryotic promoter. The genomes of the bacteria can contain modifications, such as one, two, or more modifications, selected from among: a) deletion or disruption of all or of a sufficient portion of a gene or genes, whereby the bacterium has been modified to generate penta-acylated lipopolysaccharide (LPS), wherein: the genome of the immunostimulatory bacterium is modified by deletion or disruption of all or of a sufficient portion of a gene or genes, whereby the bacterium has been modified to generate penta-acylated lipopolysaccharide; and hexa-acylated lipopolysaccharide is substantially reduced, by at least 10-fold, compared to the wild-type bacterium, or is absent; b) deletion or disruption of all or of a sufficient portion of a gene or genes, whereby the bacterium has attenuated recognition by Toll-like Receptors (TLR) 2, TLR4, and/or TLR5; c) deletion or disruption of all or of a sufficient portion of a gene or genes, whereby the bacterium does not activate the synthesis of curli fimbriae and/or cellulose; d) deletion or disruption of all or of a sufficient portion of a gene or genes, whereby the bacterium does not activate the synthesis of secreted asparaginase; e) deletion or disruption of all or of a sufficient portion of a gene or genes, whereby the bacterium is auxotrophic for purines, for adenosine, and/or for ATP; f) deletion or disruption of all or of a sufficient portion of a gene or genes, whereby the bacterium lacks flagella; g) deletion or disruption of all or of a sufficient portion of a gene or genes, whereby the bacterium has been modified to specifically infect tumor-resident myeloid cells; h) deletion or disruption of all or of a sufficient portion of a gene or genes, whereby the bacterium has been modified to specifically infect tumor-resident myeloid cells, and is unable to replicate in tumor-resident myeloid cells; and i) deletion or disruption of either or both of IppA and IppB, to decrease or eliminate lipoprotein expression in the membrane, whereby expression of an encoded therapeutic protein is increased in the tumor microenvironment and/or in tumorresident immune cells.

For example, the immunostimulatory bacteria contain modifications, including deletions, insertions, and replacements, of a), d), and f), or modifications c) and d), or modifications a), c), d), e), and f), or modifications a), c), d), e), f), and i), or modifications a), d), f), and i), or modifications c), d), and i), or modifications f) and i), or modifications a)-i), or modifications a), b), d), and f), or modifications a), b), c), and d), and other combinations of modifications a)-i). Deletion or disruption includes any modification of a gene whereby active gene product is not expressed.

In particular, provided are immunostimulatory bacteria whose genomes are modified by deletion or disruption, including by insertion, of all or of a sufficient portion of a gene or genes, whereby the bacteria have attenuated recognition by TLR2, TLR4, and TLR5. Such bacteria have low toxicity and accumulate in/colonize the tumor microenvironment and tumor-resident myeloid cells, such as macrophages. These bacteria contain plasmids that encode therapeutic products, particularly combinations of complementary products, such as a cytokine and a modified STING polypeptide, including gain-of-function/constitutively active STING proteins, STING chimeras, and chimeric STING proteins that include gain-of-function (GOF) mutations. The cytokines include, for example, IL-15/IL-15R alpha chain complex (also referred to herein as IL-15Ra/IL-15sc, or IL- 15/IL- 15Ra, or IL- 15 complex), or IL- 15, or IL- 12, or other anti -tumor immune stimulating cytokines or chemokines. The bacteria can additionally encode other products, such as anti-tumor antibodies. Combinations of products are described and provided herein. The combinations of products that stimulate or promote an anti-tumor response and/or deliver a therapeutic product, are described throughout the disclosure herein, and they are delivered by the immunostimulatory bacteria whose genomes are modified so that the bacteria have low toxicity and effectively colonize tumors, the tumor microenvironment, and/or tumor-resident immune cells, such as macrophages. Exemplary of such bacteria are those of species, such as Salmonella, Listeria, and Escherichia, that are modified so that they do not have flagella, and are modified so that they contain lipopolysaccharide (LPS) with penta-acylated lipid A, such as by rendering the bacteria msbB'i agP. The bacteria additionally can be modified by elimination of curli fimbriae and/or have reduced or eliminated cellulose production and biofilm formation, such as by modifying the bacteria so that they are csgD'. It is shown herein that bacteria with these modifications have no maximum tolerated dose (MTD), and exhibit high tumor colonization.

In all embodiments, the immunostimulatory bacteria also can comprise or further comprise deletion of or disruption of the genes encoding the flagella, whereby the bacterium is flagellin" (such as, for example, fliCIfljB' in Salmonella) and does not produce flagella (i.e., can be referred to as flagellin deficient or flagellin"), where the wild-type bacterium has flagella. The immunostimulatory bacteria can be auxotrophic for purines, such as auxotrophic for adenosine; or auxotrophic for adenosine, adenine, and/or ATP. The immunostimulatory bacteria also can be purL. The immunostimulatory bacteria also can be pagP. The immunostimulatory bacteria also can be aspartate-semialdehyde dehydrogenase" (asd'), such as where the bacterium is asd' by virtue of disruption or deletion of all or a portion of the endogenous gene encoding aspartate-semialdehyde dehydrogenase (asd), whereby endogenous asd is not expressed. The bacteria can encode aspartate-semialdehyde dehydrogenase (asd) on the plasmid under control of a bacterial promoter. The immunostimulatory bacteria also can be msbB', or can be pagP' ImsbB' . For example, the immunostimulatory bacteria can be asd , purl', msbB', flagellin" (such as fliC/fljB'), and pagP, or they can be asd', csgD' , purl' , msbB', flagellin" (such as fliC'lfljB'), and pagP. In some embodiments, the immunostimulatory bacteria are ansB', asd', csgD', purl', msbB', flagellin" (such as fliCIfljB ), and pagP.

Provided are immunostimulatory bacteria that contain a plasmid encoding a therapeutic product under control of a eukaryotic promoter, or that encode a plurality of products under control of a plurality of eukaryotic promoters, or under control of a single promoter. The genome of the immunostimulatory bacteria is modified by deletion of a sufficient portion of a gene or genes, or by the disruption of a gene or genes, whereby the bacterium is one or more of ansB', asd, csgD', purl', msbP,

- 31 -

RECTIFIED SHEET (RULE 91 ) ISA/EP flagellin" (such as fliC'/fljB ), and pagP'. The immunostimulatory bacteria provided herein also include those that have the genes IppA (IppB) and/or IppB (lpp2 which encode major outer membrane lipoproteins Lppl (LppA) and Lpp2 (LppB), respectively, deleted or disrupted, to eliminate or substantially reduce expression of the encoded lipoprotein(s). In particular, the immunostimulatory bacteria are IppA' and IppB'. Provided are immunostimulatory bacteria that contain a plasmid encoding an anti-cancer therapeutic, or an anti-pathogen therapeutic, under control of eukaryotic regulatory sequences, and that are IppA' and IppB'. For example, the immunostimulatory bacteria can be ansB', asd, csgD' , purl' , msbB', flagellin" (such as fliC~/fljB~), pagP', IppA', and/or IppB' .

In embodiments herein, the therapeutic product is an anti-cancer therapeutic or a therapeutic used in cancer therapy. The encoded product(s) can be operably linked to nucleic acid encoding a secretion signal, whereby, when expressed, the therapeutic product is secreted, such as secreted from a tumor-resident immune cell.

Any of the immunostimulatory bacteria also can have one or more genes or operons, involved in Salmonella pathogenicity island 1 (SPI-1) invasion, deleted or inactivated, whereby the immunostimulatory bacteria do not invade or infect epithelial cells. For example, the one or more genes/operons are selected from among avr.A, hilA, hilD, invA, invB, invC, invE, invF, invG, invH, invl, invJ, iacP, iagB, spaO, spaQ, spaR, spaS, orgA, orgB, orgC, prgH, prgl, prgj, prgK, sicA, sicP, sipA, sipB, sipC, sipD, sirC, sopB, sopD, sopE, sopE2, sprB, and sptP.

The plasmid in the immunostimulatory bacteria can be present in low copy number or medium copy number. The plasmid can contain a medium-to-low copy number origin of replication, such as a low copy number origin of replication. The plasmid may be present in medium-to-low copy number depending on the ORI sequence, the size of the plasmid, and culturing conditions. In some embodiments, the plasmid is present in higher copy number. Generally, medium copy number is less than 150 or less than about 150, and more than 20 or about 20, or is between 20 or 25 and 150; and low copy number is less than 25, or less than 20, or less than about 25, or less than about 20 copies. In particular, low to medium copy number is less than about 150 copies, or less than 150 copies; low copy number is less than about 25 copies, or less than 25 copies.

Encoded therapeutic products include nucleic acids and proteins. The plasmid can encode two or more therapeutic products. Exemplary products include, but are not limited to, a cytokine, a protein that constitutively induces a type I IFN, and a costimulatory receptor or ligand. Further exemplary combinations are described below. In some embodiments, the co-stimulatory molecule lacks all or a portion of the cytoplasmic domain for expression on an antigen-presenting cell (APC), whereby the truncated molecule is capable of constitutive immuno-stimulatory signaling to a T-cell through co-stimulatory receptor engagement, and is unable to counter-regulatory signal to the antigen-presenting cell (APC), due to the deleted cytoplasmic domain or deleted portion thereof.

The encoded therapeutic products can be operatively linked to nucleic acid encoding regulatory sequences recognized by a eukaryotic host, such as, for example, secretion signals, to effect secretion from a cell comprising the bacterium or plasmid. In embodiments where the immunostimulatory bacteria encode two or more products, expression of each product can be under control of a separate promoter. Alternatively, two or more products can be expressed under control of a single promoter, and each product is separated by nucleic acid encoding, for example, an internal ribosomal entry site (IRES), or a 2A peptide, to effect separate expression of each encoded therapeutic product. Exemplary 2A peptides are T2A, F2A, E2A, or P2A, which can flank nucleic acids encoding the therapeutic products, to effect separate expression of the therapeutic products expressed under control of a single promoter. The therapeutic products are expressed under control of a eukaryotic promoter, such as an RNA polymerase (RNAP) II promoter, or an RNA polymerase III promoter. These include an RNA polymerase II promoter that is a viral promoter, or a mammalian RNA polymerase II promoter, such as, but not limited to, a cytomegalovirus (CMV) promoter, an SV40 promoter, an Epstein-Barr virus (EBV) promoter, a herpesvirus promoter, an adenovirus promoter, an elongation factor- 1 alpha (EF-la) promoter, a UBC promoter, a PGK promoter, a CAGG promoter, an adenovirus 2 or 5 late promoter, an eIF4Al promoter, a CAG promoter, or a CD68 promoter. The plasmids further can include other eukaryotic regulatory sequences, such as terminators and/or promoters, selected from among SV40, human growth hormone (hGH), bovine growth hormone (bGH), MND (a synthetic promoter that contains the U3 region of a modified MoMuLV LTR with myeloproliferative sarcoma virus enhancer), chicken beta-globulin, and rbGlob (rabbit globulin) genes, to control expression of the therapeutic product(s). Other regulatory sequences include a poly(A) tail, a Woodchuck Hepatitis Virus (WHP) Posttranscnptional Regulatory Element (WPRE), and a Hepatitis B virus Posttranscriptional Regulatory Element (HPRE).

Immunostimulatory bacteria and vaccines

Provided are immunostimulatory bacteria that comprise a plasmid that encodes a therapeutic product; the bacteria comprise genome modifications, such as insertions, deletions, replacements, transposons, whereby the bacterium does not produce active thymidylate synthase, and requires supplementation for growth. Supplementation includes nutrients that can bypass the reactions catalyzed by thymidylate synthase so that the bacteria can replicate. Supplementation includes one or more of thymine, thymine derivatives, thymidine, thymidine derivatives, thymine precursor(s), thymidine precursor(s), or thymidine monophosphate precursor(s). The bacteria can additionally comprise additional modifications reduce or eliminate activation of TLR2, and optionally TLR4 and/or TLR5 in a host, such as a human or other mammal.

Also provided are immunostimulatory bacteria that comprise a plasmid that encodes a therapeutic product, where the bacterium comprises genome modifications, whereby the bacterium does not secrete active asparaginase; and the bacterium comprises genome modifications that reduce or eliminate activation of TLR2, and optionally TLR4 and/or TLR5 in a host.

These immunostimulatory bacteria additionally can include genome modifications, by deletion or disruption or modification of all or of a sufficient portion of the gene ansB encoding L-asparaginase II, whereby the bacterium is ansB' and does not express active L-asparaginase II.

Also provided are immunostimulatory bacterium, comprising genome modification(s) that reduce or eliminate activation of TLR2, whereby induction of type I interferon (IFN) is not inhibited by TLR2, where: the immunostimulatory bacterium comprises genome modification(s) whereby it cannot replicate in vivo, but can replicate when grown in vitro with nutritional supplementation; and the genome modification(s) eliminate(s) or inactivate(s) thymidylate synthase, whereby the bacterium is thyA~ and/or asd' or both.

It is described herein that activation of TLR2 can inhibit induction of type I IFN. It is shown herein that expression of a protein, such as a STING protein by a bacterium or to delivery vehicle that activates TLR2, or TLR4/5 and TLR2, is not an advantageous combination since activation of the TLRs, such as TLR2, inhibits type I interferon. Type I IFN is, for example, an interferon-a and/or interferon-p.

Provided are immunostimulatory bacteria that comprise genome modifications whereby activation of TLR2 is reduced or eliminated, whereby induction of type I IFN is not inhibited by TLR2, wherein the immunostimulatory bacterium comprises a plasmid that encodes an interferon, or encodes a modified STING protein that constitutively induces type I interferon, and encodes an antigen or protein from a pathogen or tumor. These bacteria also can include genome modifications whereby TLR4 and/or TLR5 activation/induction is reduced or eliminated.

Provided are immunostimulatory bacteria that comprise a plasmid encoding a therapeutic product, where: the genome of the immunostimulatory bacterium is modified by deletion or disruption of all or of a sufficient portion of a gene or genes, whereby the bacterium has been modified to generate lipopolysaccharide (LPS) with penta-acylated lipid A; lipopolysaccharide with hexa-acylated lipid A is substantially reduced, by at least 10-fold, compared to the wild-type bacterium, or is absent; the genome of the bacterium is modified, whereby the bacterium itself does not inhibit or prevent induction of type I interferon (IFN) in an infected immune cell; and the genome of bacterium is modified to be auxotrophic for an essential nutrient. For example provided are immunostimulatory bacterium with genome modifications, whereby the bacterium does not encode or produce active asparaginase and/or thymidylate synthase. Any of the immunostimulatory bacteria provided herein can have genome modifications that comprise deletions, insertions, and/or replacements whereby the bacterium is thyA' and/or asd or both thyA~ and as . Additionally, in some embodiments in which the bacteria are asd by virtue of genome modifications, the bacteria can include nucleic acid encoding asd on the plasmid such that it is expressed in vivo. The particular embodiments and applications for particular auxotrophies and complementation on the plasmid are described in the detailed description and/or known to those of skill in the art.

Also provided are immunostimulatory bacteria, comprising a plasmid encoding a therapeutic product, where: the genome of the immunostimulatory bacterium is modified by deletion or disruption or translocation of all or of a sufficient portion of a gene or genes, whereby the bacterium has been modified to generate lipopolysaccharide (LPS) with penta-acylated lipid A; lipopolysaccharide with hexaacylated lipid A is substantially reduced, by at least 10-fold, compared to the wild- type bacterium, or is absent; and the genome of bacterium is modified whereby it does not produce active thymidylate synthase, whereby the bacterium is thyA'.

Also provided are immunostimulatory bacteria that comprise a plasmid encoding a therapeutic product, where: the genome of the immunostimulatory bacterium is modified by deletion or disruption of all or of a sufficient portion of a gene or genes, whereby the bacterium lacks flagella; the unmodified immunostimulatory bacterium has flagella; and the genome of the bacterium is modified, whereby it does not produce active thymidylate synthase.

Any of the immunostimulatory bacteria provided herein can have genome modifications, such as modifications that render the bacteria csgD', whereby the bacteria lack curli fimbriae. Any of the immunostimulatory bacteria can comprises genome modifications that reduce or eliminate activation of TLR4 and/or TLR5. The immunostimulatory bacteria provided herein can comprise genome modifications that result in a bacterium that does not have flagella; and the wild-type of the bacterium has flagella. These bacteria further can have genome modification whereby they do not produce curli fimbriae. They also can comprise genome modifications that result in penta-acylated lipopolysaccharides. Thus, the bacteria provided herein can lack flagella and be msbB'/pagP~.

Provided herein are immunostimulatory bacteria, comprising genome modifications that reduce or eliminate activation of TLR2, whereby induction of type I IFN is not inhibited by TLR2, where: the immunostimulatory bacterium can replicate in vivo in a eukaryotic host; and the immunostimulatory bacterium comprises a plasmid that encodes a therapeutic product or products that include tumor-associated antigen, or encodes a tumor antigen and a product that is part of a cytosolic DNA/RNA sensor pathway that leads to expression of type I interferon (IFN), or the encoded product is a type I IFN that is interferon-alpha or is interferonbeta, or the encoded product is both IFN-alpha and IFN-beta, such as a STING protein, or encodes a tumor-associated antigen and interferon alpha, or encodes a tumor-associated antigen and interferon beta. Among the products that are part of the cytosolic DNA/RNA sensor pathway, such as modified protein that results in increased induction of type I interferon, such as a STING protein that is a modified STING protein that has increased induction of type I interferon compared to the unmodified human STING protein. Modified STING proteins include, for example any described below, that have result in increased or constitutive expression or

- 36 -

RECTIFIED SHEET (RULE 91) ISA/EP induction of type I interferon, particularly compared to unmodified human STING protein. The modified STING proteins additionally can have reduced NF-KB signaling compared to wild-type human STING protein. Included are chimeric STING proteins as described herein (see, also International PCT Publication WO 2020/176809, and US Publication No. 2020/027061). For example, the STING protein is one that comprises replacements corresponding to N154S, R284G, or N154S/R284G, with reference to a human STING protein. Exemplary immunostimulatory bacteria are those that encode a modified STING protein that constitutively induces type I interferon and a tumor antigen and/or a cytokine, such as an IL-15 receptor complex. The immunostimulatory bacteria have genome modifications whereby the bacteria is/are flagellin deficient so that they do not have flagella and have penta-acylated LPS, such as by virtue of genome modifications that render the bacteria msbR/pagP' . The bacteria also can lack curli fimbriae, such as by virtue of genome modifications that render them csgD'. The bacteria optionally are auxotrophic for a required nutrient, such as bacteria that are thyA' and/or adenosine auxotrophs, or other such auxotrophy. The bacteria also can be ansR. The particular combination of genome modifications, as described herein, depends upon the intended use of the bacteria and desired effects as does the particular selection of encoded therapeutic products. As described herein the promoters and other regulatory sequences that control expression of the encoded products also depends upon the intended use of bacteria. As described throughout the disclosure herein, the therapeutic proteins encoded on the plasmids can be under control of eukaryotic regulatory sequences, such as, for example, for antitumor therapy embodiments, or can be under control of bacterial promoters for embodiments, such as, for example, as certain vaccines in which the encoded proteins or RNA are intended for delivery

With respect to encoded products, that immunostimulatory bacteria can encode a therapeutic product that is part of a cytosolic DNA/RNA sensor pathway that leads to expression of type I interferon (IFN), or the encoded product is interferon-alpha or is interferon-beta, or the encoded product is both IFN-alpha and IFN-beta. Other products include cytokines, antibodies, bi-specific engager antibodies, and tumor antigens, or other antigens, such as pathogen antigens for vaccination. Exemplary of a therapeutic product that is part of a cytosolic DNA/RNA sensor pathway that leads to expression of type I interferon (IFN) is a STING protein, particularly one that is modified to have increased, particularly, constitutive activity, so that sensing of cytosolic DNA/RNA or the presence of such DNA/RNA is/are not required, nor is any ligand for such pathway. The plasmid can encode an antigen, epitope(s), or protein from a pathogen or a tumor.

Provided are immunostimulatory bacteria that comprise a plasmid encoding a combination of heterologous products, where: the genome of the immunostimulatory bacterium is modified by deletion or disruption of all or of a sufficient portion of a gene or genes, whereby the bacterium has attenuated recognition by TLR2, and optionally one or both of TLR4 and TLR5; one product is part of a cytosolic DNA/RNA sensor pathway that leads to expression of type I interferon (IFN); and a second product that is an antigen, an epitope or epitopes from is antigen, or is protein for immunization against the pathogen or tumor. In some embodiments, the genome of the bacterium is modified whereby the bacterium does not have curli fimbriae.

Also provided are immunostimulatory bacteria, comprising nucleic acid operatively linked to a prokaryotic promoter, particularly where the nucleic acid is encoded on plasmid, where: the nucleic acid, which is expressed in the bacteria under control of the prokaryotic promoter, encodes RNA that lacks sequences necessary for translation by a prokaryote, whereby the RNA is produced in the bacterium. Exemplary of such are embodiments where the encoded RNA lacks a Shine-Dalgarno sequence, and/or comprises an Internal Ribosome Entry Site (IRES). The nucleic acid also can encode a translational read through 2A peptide so that discrete products are produced upon expression of the nucleic acid when the nucleic acid encodes a polycistronic message.

Provided are immunostimulatory bacteria, comprising nucleic acid, such as on a plasmid, operatively linked to a prokaryotic promoter, where the nucleic acid comprises RNA that lacks sequences necessary for translation by a prokaryote. As a result the bacteria can transcribe, but not translate the encoded RNA. The bacteria can then be used as RNA delivery vehicles. As noted, this can be achieved where the encoded RNA lacks a Shine-Dalgarno sequence. For polycistronic nucleic acid, the nucleic acid can comprise a 2A peptide, such as one or more of T2A, P2A, E2A, or F2A.

Provided are immunostimulatory bacteria, where the nucleic acid encoding a therapeutic product is operatively linked to a prokaryotic promoter; the nucleic acid encodes RNA that lacks sequences necessary for translation by a prokaryote, whereby the RNA is produced in the bacterium; the RNA lacks a Shine-Dalgarno sequence, and comprises an Internal Ribosome Entry Site (IRES), and/or can also include a translational read through 2A peptide. In some embodiments, The nucleic acid encoding the therapeutic product(s) can be operably linked to nucleic acid encoding a secretion signal, whereby, when expressed, the therapeutic product(s) is/are secreted.

Bacterial promoters for expression of encoded therapeutic products include any recognized by the bacterial or bacteriophage RNA polymerase, such as a bacterial promoter or a bacteriophage promoter. Where the promoter is one only recognized by a bacteriophage RNA polymerase, the bacteria can encode the bacteriophage polymerase, such as a T7 RNA polymerase. Exemplary promoters are any comprising any of SEQ ID NOs: 393-396, respectively: attatgtcttgacatgtagtgagtgggctggtataatgcagcaag, or ttatgcttgacgctgcgtaaggtttttgttataatacaccaag, or attatgtcttgacatgtagtgagtgggctggtaaatgcagcaag, or gatcccggagttcatgcgtgatgcaatgaaagtgccgttctacttcggtgggacctcact gcttatcgttgttgtcgtgattatggact ttatggctcaagtgcaaactctgatgatgtccagtcagtatgagtctgcattgaagaagg cgaacctgaaaggctacggccgttaa ttggtcgcctgagaagttacggagagtaaaaatgaaagttcgtgcttccgtcaagaaatt atgccgtaactgcaaaatcgttaagc gtgatggtgtcatccgtgtgatttgcagtgccgagccgaagcataaacagcgccaaggct gattttttcgcatatttttcttgcaaag ttgggttgagctggctagattagccagccaatcttttgtatgtctgtacgtttccatttg agtatcctgaaaacgggcttttcagcatgg tacgtacatattaaatagtaggagtgcatagtggcccgtatagcaggcattaacattcct gatcagaaacacgccgtgatcgcgtta acttcgatctacggtgtcggcaagacccgttctaaagccatcctggctgcagcgggtatc gctgaaaatgttaagatcctctagatt taagaaggagatatacat (Salmonella rpsm promoter).

The immunostimulatory bacteria include genome modifications, such as elimination of flagella and/or other modifications so that the bacteria do not infect epithelial cells, but still infect or accumulate in or preferentially infect (infect to a greater extent or amount than the unmodified bacteria, and infect other cells types to lesser extent or amount than the unmodified bacteria) phagocytic cells, such as tumorresident myeloid cells in subjects with tumors, and tissue-resident myeloid cells, such at or near the site of vaccination when the bacteria are vaccines. Such bacteria include those that contain genome modifications so that, for bacteria that have flagella, the modifications render the bacterial flagellin deficient so that the bacteria lack flagella.

Immunostimulatory bacteria provided herein include those that comprise a plasmid encoding a therapeutic product, where infection of a macrophage by the bacterium converts an macrophage to an Ml or Ml -like phenotype macrophage. The encoded therapeutic product can be a therapeutic product that is part of a cytosolic DNA/RNA sensor pathway that leads to expression of type I interferon (IFN), such as one that modified so that expression of the type I IFN is constitutive The therapeutic product can be one that is a gain-of-function (GOF) variant of the therapeutic product that is part of the cytosolic DNA/RNA sensor pathway, wherein the variant GOF product does not require cytosolic nucleic acids, nucleotides, dinucleotides, or cyclic dinucleotides to result in expression of type I IFN. Exemplary of such products is a modified or variant STING protein. Infection by such bacteria can convert human M2 macrophages into Ml -like type I IFN producing cells. Exemplary of such bacteria are those that lack flagella, where the wild-type bacterium has flagella, and are pagP~ ImsbB'. Thus, provided are methods for converting an M2 macrophage into one with an Ml or Ml -like phenotype by introducing or infecting the M2 macrophage with a immunostimulatory bacterium that lacks flagella and has penta-acyl ated LPS, such as a bacterium that is msbB' Ipag , and that encodes a be a therapeutic product that is part of a cytosolic DNA/RNA sensor pathway that leads to expression of type I interferon (IFN), such as one that modified so that expression of the type I IFN is constitutive, such as the variant or modified STING proteins as described herein, including those that result in constitutive type I IFN expression. Exemplary of such bacteria are those that are a Salmonella strain species or type.

The immunostimulatory bacteria can encode therapeutic products on the plasmid, such therapeutic products include an anti-cancer therapeutics, which include any product that is used for or in connection or conjunction with a cancer treatment. The therapeutic products also include products that are used for or in connection or conjunction with treatments for a pathogen, such as a viral, bacterial, yeast, or parasitic pathogen. The products can be an anti-viral therapeutics and/or an anti- pathogenic bacterial therapeutics. It is understood that some therapeutic products are used for treatment of a variety of indications; for example, an anti-cancer product can also be effective or used for treatment of viral infections. Anti-virals include vaccines, and therapeutic products that inhibit a viral enzyme or inhibit viral replication. Therapeutics include the anti-viral therapeutics, such as a viral antigen whose expression results in an immune-protective response against a virus, and antibodies that binds to and/or interact with a viral antigen, whereby a virus is inhibited or blocked, or anti-viral immunity results. Other therapeutics are anti -bacterial therapeutics, such as bacterial antigens whose expression results in an immune- protective response against the bacterial pathogen, and/or antibodies that bind to or interact with a bacterial antigen, whereby the pathogenic bacterium is inhibited or blocked, or anti-pathogenic bacterium immunity results. Antivirals encoded by the bacteria include anti-viral therapeutics for treating a virus or infectious agent that causes persistent infection. Exemplary of anti-viral therapeutics are viral antigens or epitopes of an antigen, such as, but not limited to, a viral surface protein, or a viral nucleocapsid protein, or a viral nonstructural protein, or a virus open reading frame protein, such as, for example, embodiments in which a therapeutic product is a viral surface antigen or portion thereof sufficient to produce an immune response in a host, embodiments in which the therapeutic product interferes with viral gene expression or replication.

The vims or other infectious agent or pathogen can be one that causes chronic infection, and/or latent infection, and/or slow infection. Exemplary of viral pathogens are a vims or infectious agent selected from among T-Cell leukemia vimses, Epstein- Barr vims, cytomegalovims, herpesviruses, varicella zoster vims, measles vims, papovavimses, prions, hepatitis vims type A, B, C, D and E, adenovimses, parvoviruses, human immunodeficiency vims (HIV), coronavimses, smallpox vims, poliovims, influenza vims, rotavims, yellow fever vims, mumps vims, rubella vims, and papillomaviruses, such as a HIV or hepatitis vims. Other infectious agents include prions and protozoans.

Therapeutic products encoded by the immunostimulatory bacteria include immunostimulatory proteins, such as the aforementioned Stimulator of Interferon Genes (STING) protein, a modified STING protein, a cytokine, a chemokine, or a costimulatory receptor or ligand. The immunostimulatory bacteria include those that comprise a genomic modification whereby the bacteria lack flagella, and/or are pagP~ or msbR IpagP . The products include immunostimulatory proteins that confer or contribute to anti-tumor immunity in the tumor microenvironment is a cytokine or a chemokine.

The immunostimulatory bacterium include any described herein and that also compnse genomic modification whereby they do not express asparaginase or activate the synthesis of secreted asparaginase; and/or the genome of the immunostimulatory bacterium is modified by deletion or disruption of all or of a sufficient portion of the gene ansB encoding L-asparaginase II, whereby the bacterium is ansB' and does not express active L-asparaginase II. Such bacteria can encode any therapeutic product of interest, including any provided or described herein, whereby the resulting bacterium is an anticancer therapeutic that colonizes tumors and/or the tumor microenvironment, whereby the ansB' phenotype reduces or eliminates production of active asparaginase. Provided herein are immunostimulatory bacteria, comprising nucleic acid operatively linked to a prokaryotic promoter, where: the nucleic acid comprises RNA that lacks sequences necessary for translation by a prokaryote, whereby the RNA is produced in the bacterium, but cannot be translated by the bacterium; the bacterium has genomic modifications whereby infection is restricted to myeloid cells; and the RNA encodes a therapeutic product or is a therapeutic product.

Provided herein is an RNA delivery system, comprising an immunostimulatory bacterium that primarily or solely infects myeloid cells and that comprises RNA encoded by the bacterium under control of a prokaryotic promoter, where: the RNA lacks regulatory sequences necessary for translation by the bacterium; and the RNA encodes a therapeutic product or is a therapeutic product. As discussed, the transcribed RNA lacks a Shine-Dalgamo sequence or include or lack other sequences so that it is not translated by bacterial ribosomes, but is translated by eukaryotic ribosomes in a host. The RNA can comprise a Kozak consensus sequence, such as, for example, where a Kozak consensus sequence is ACCAUGG (SEQ ID NO: 397). The immunostimulatory bacteria of includes those that lack flagella and are msbB'/pagP'. In the RNA delivery systems, the bacteria can contain a plasmid that encodes the therapeutic product or products. In some embodiments, the nucleic acid encoding the therapeutic product(s) is operatively linked to a prokaryotic promoter that is inducible or one that is constitutive. The encoding nucleic acid can comprise nucleic acid encoding an Internal Ribosome Entry Site (IRES) or other sequence so that the transcribed RNA is not translated by bacterial ribosomes but is translated by eukaryotic ribosomes. As a result the bacteria encode and produce RNA encoding any therapeutic protein, such as an antigen and other payload, but does not translate the RNA. The RNA is translated after administration of the bacteria to a host, such as a human, where, for example, the enter phagocytic cells, where the RNA can be translated.

Provided are immunostimulatory bacteria and RNA delivery system where the genome of the immunostimulatory bacterium is modified by deletion or disruption or other change of all or of a sufficient portion of the gene ansB, encoding L-asparaginase II, whereby the bacterium is ansB' and does not express active L-asparaginase II. The immunostimulatory bacteria and RNA delivery systems can further contain modification(s) of the genome, such as by deletion or disruption or mother change of all or of a sufficient portion of the gene csgD, whereby the bacterium is ansB' and does not express active L-asparaginase II, and is csgD' and does not activate the synthesis of curli fimbriae and/or the bacteria can include further modification(s) of the genome whereby biofilm formation is impaired. These bacteria and RNA delivery systems can further include genome modifications whereby the bacteria is flagellin" and does not produce flagella, where the wild-type bacterium has flagella. Hence, for example, the immunostimulatory bacterium or RNA delivery system has genomes modifications, whereby the bacterium is csgD'hnsbB' IpagP'’ and the bacterium or RNA delivery system also can include genome modifications, whereby the bacterium lacks flagella. Other genome modifications can be included, whereby the bacterium lacks flagella, and is IppA' UppB , and optionally is csgD . Any of the immunostimulatory bacteria and RNA systems described herein can be auxotrophic for a nutrient, such as purines, such as auxotrophy for adenosine and/or for all or any of adenosine, adenine, and ATP. Adenosine auxotrophy is advantageous for bacteria that accumulate in the tumor microenvironment or tumor-resident macrophage; adenosine accumulation occurs in the tumor microenvironment. The bacteria provided herein can include additional genome modifications, including those whereby the is purl', including by complete deletion of the gene, and/or is pagP' and/or is asd' or thyA' or both.

The immunostimulatory bacteria provided herein can be aspartate-semialdehyde dehydrogenase" (asd'), where the bacterium is asd' by virtue of disruption of or deletion or transposition or other modification of all or a portion of the endogenous gene encoding aspartate-semialdehyde dehydrogenase (asd), whereby endogenous asd is not expressed or functional enzyme is not produced, or is thyA~ by virtue of the disruption of or deletion of all or a portion of the endogenous gene or genes, whereby endogenous thymidylate synthase is not expressed or functional enzyme is not produced. Hence provided are immunostimulatory bacteria that are aspartate-semialdehyde dehydrogenase" (asd'), where: the bacterium is asd' by virtue of disruption of or deletion of all or a portion of the endogenous gene encoding aspartate-semialdehyde dehydrogenase (asd), whereby endogenous asd is not expressed or functional enzyme is not produced; and the bacterium is thyA' by virtue of the disruption of or deletion of all or a portion of the endogenous gene or genes, whereby endogenous thymidylate synthase is not expressed or functional enzyme is not produced. The immunostimulatory bacteria can encodes aspartate- semialdehyde dehydrogenase (asd) on the plasmid under control of a bacterial promoter so that the asd can be produced in vivo. Exemplary of the immunostimulatory bacteria provided are those here the unmodified bacterium is a Salmonella bacterium.

The immunostimulatory bacterium of provided herein can be msbB' by virtue of genome modifications, including, but not limited to, complete or partial deletion of the gene locus. Full deletion results in bacteria that grow better than those that retain part of the gene.

The bacteria provided herein are immunostimulatory bacteria that are asd', purl', msbB', flagellin" and pagP' or that any immunostimulatory bacteria described above or herein that are asd', csgD',purI', msbB', flagellin" and pagP'; or thyA', csgD' , purP , msbB', flagellin" and pagP'; or ansB', asd', csgD', purP, msbB', flagellin" and pagP'; or ansB', thyA', csgD', purP, msbB', flagellin" and pagP', or is ansB', thyA', csgD', purP, msbB', flagellin" and pagP'.

The immunostimulatory bacteria encode therapeutic products, such as but not limited to, anti-cancer therapeutics and/or therapeutics for treating diseases, disorders, and conditions caused by pathogens or other diseases, disorders, and conditions. Exemplary encoded products include, but are not limited to combinations of products, such as a modified STING and IL- 15 or IL- 15/IL- 15R alpha chain complex, where the STING constitutively induces type I LFN in the absence of cGAS and/or any STING ligands.

Provided are the immunostimulatory bacteria as described herein that encode a therapeutic product that is an anti-viral product, such as, for example, a viral antigen and/or an anti-viral antibody. Viruses include, for example, viruses that cause chronic infections or latent infections, such as, but not limited to, a hepatitis virus, a herpes virus, a varicella zoster virus, a poxvirus, a measles virus, and a retrovirus.

In the immunostimulatory bacterium the copy number of the plasmids is from low to high. In some embodiments, the copy number is from low to medium, such as where number of copies of the plasmid is less than 150. In other embodiments the copy number is from 150 or is greater than 150. Hence in embodiments, the number of copies of the plasmid is 150 copies or fewer, or is less than or equal to 150. In other embodiments, the plasmid is present in low copy number, and low copy number is less than 25 or less than 20 or less than about 25 or less than about 20 copies, typically less than 25.

The encoded therapeutic products include, proteins and also nucleic acids, such as RNA products, antigens, antibodies. The plasmid can encode two or more products. Exemplary products are any that are used in the treatment of cancer. Also included are produces used as anti-viral treatment(s). For example, the bacteria can encode two or more products selected from among a cytokine, a protein that constitutively induces a type I IFN, and a co-stimulatory receptor or molecule. The co-stimulatory molecule can be modified so that it lacks a cytoplasmic domain. The products can have complementary activities; in some embodiments the activities are synergistic. The nucleic acid encoding one or more of the therapeutic product or products comprises nucleic acid can encode a signal for secretion of the therapeutic product(s) from a cell comprising the bacterium. The nucleic acid encoding the product on the plasmid can be operatively linked to regulatory sequences recognized by a eukaryotic host. In embodiments, in which the immunostimulatory bacterium encodes two or more products, expression of each product can be under control of a separate promoter, or expression of all two or more can be under control of a single promoter, such as where nucleic acid encoding each product is separated by nucleic acid encoding a 2A peptide to effect separate translation of each encoded therapeutic product. Exemplary 2A peptides, include T2A, F2A, E2A, or P2A peptide, which effect separate expression of therapeutic products expressed under control of a single promoter. Eukaryotic regulatory signals can control expression of the produce or products. Eukaryotic promoters include RNA polymerase II promoters and RNA polymerase III promoters. Eukaryotic RNA polymerase II promoters include viral promoters from viruses that infect eukaryotes, and mammalian RNA polymerase II promoters. Exemplary promoters include, but are not limited to, viral promoters, such as a cytomegalovirus (CMV) promoter, an SV40 promoter, an Epstein Barr virus (EBV) promoter, a herpes virus promoter, and an adenovirus promoter; an elongation factor- 1 (EF-1) alpha promoter, or an MND promoter, or a UBC promoter, or a PGK promoter, or a CAG promoter, such as an EF-1 alpha, an adenovirus 2 or 5 late, a CMV, an SV40, an MND, a PGK, an EIF4A1, a CAG, or a CD68 promoter. Viral promoters include those that are late promoters. The immunostimulatory bacteria include those where the plasmid comprises regulatory sequences that comprise a terminator and/or promoters) selected from among SV40, hGH, BGH, MND, chicken beta-globulin, and rbGlob (rabbit globulin) genes, to control expression of the therapeutic product(s). The encoded therapeutic product(s) can be operatively linked to a signal sequence for secretion from a cell containing the plasmid; or in some embodiments can be designed or modified to be expressed on the surface of the cell in which they are produced. The plasmid that encodes the therapeutic product(s) can comprises a nucleic acid construct that includes an enhancer, a promoter, the open reading frame encoding the therapeutic product or heterologous protein, and a polyA tail. Exemplary of the plasmids in the bacteria are those where the plasmid comprises a construct that includes an enhancer, a promoter, an IRES, the open reading frame encoding the therapeutic product or heterologous protein, and a polyA tail, and those where the plasmid comprises a construct that includes an enhancer, a promoter, an IRES, a localization sequence, the open reading frame encoding the therapeutic product or heterologous protein, and a polyA tail. The constructs can include posttranscriptional regulatory elements, such as a Woodchuck Hepatitis Virus (WHP) Posttranscnptional Regulatory Element (WPRE), or a Hepatitis B virus Posttranscriptional Regulatory Element (HPRE).

The bacteria provided herein contain plasmids that encode a therapeutic product that is part of a cytosolic DNA/RNA sensor pathway that leads to expression of type I interferon (IFN), or a variant of the therapeutic product. These products can be modified to have increased or constitutive activity for expression of type I IFN. These products in unmodified form sense or interact directly or indirectly with cytosolic nucleic acids, nucleotides, dinucleotides, or cyclic dinucleotides, to induce expression of type I IFN, and the variant or modified protein induces expression of type I IFN in the absence of the sensing or interacting with the cytosolic nucleic acids, nucleotides, dinucleotides, or cyclic dinucleotides so that in the therapeutic product is a variant that, when expressed in a subject, leads to constitutive expression of type I IFN. Mutations in the variant proteins include those that result in a gain-of-function so that the variant that does not require cytosolic nucleic acids, nucleotides, dinucleotides, or cyclic dinucleotides, or ligands, to result in expression of type I IFN. These products that are part of a cytosolic DNA/RNA sensor pathway that leads to expression of type I interferon (IFN) include, but are not limited to, STING, RIG-I, MDA-5, IRF-3, IRF-5, IRF-7, IRF-8, TRIM56, RIP1, Sec5, TRAF3, TRAF2, TRAF6, STAT1, LGP2, DDX3, DHX9, DDX1, DDX9, DDX21, DHX15, DHX33, DHX36, DDX60, and SNRNP200, such as STING, RIG-I, IRF-3, IRF- 5, IRF-8, or MDA5, particularly variants of these proteins that have increased activity, or that results in constitutive expression of type I interferon (IFN). Mutations include those that, in humans, promotes or causes interferonopathies. Other mutations in these proteins include mutations that eliminate a phosphorylation site in the protein to thereby reduce nuclear factor kappa-light-chain-enhancer of activated B cell (NF-KB) signaling, and combinations of mutations. Provided are immunostimulatory bacteria, where the therapeutic product is a variant thereof that has increased activity or constitutive activity; and the therapeutic product is STING, RIG-I, IRF-3, IRF-5, IRF-8, or MDA5, such as, for example, where the therapeutic product is a variant of STING, RIG-I, IRF-3, IRF-5, IRF- 8, or MDA5 that comprises a gain-of-function mutation resulting in increased or constitutive expression of type I IFN, and optionally mutations or replacements of the C- terminal tail (CTT) resulting in decreased NF-KB signaling activity, such as where the therapeutic product is a variant of STING, RIG-I, IRF-3, IRF-5, IRF-8, or MDA5 in which one or more serine (S) or threonine (T) residue(s) that is/are phosphorylated as a consequence of viral infection, is/are replaced with an aspartic acid (D), whereby the resulting variant is a phosphomimetic that constitutively induces type I IFN. The therapeutic product is IRF-3 that has one or more replacement(s) at residues at positions 396, 398, 402, 404 and 405, with reference to SEQ ID NO:312; and the residues are replaced with aspartic acid residues, such as an IRF-3 that compnses the replacement S396D with reference to SEQ ID NO:312, such as, for example, where IRF- 3 comprises the replacements S396D/S398D/S402D/T404D/S405D with reference to SEQ ID NO: 312. Other examples, include where the therapeutic product that senses cytosolic DNA/RNA is a variant STING, MDA5, RIG-I or IRF-3; and unmodified STING has the sequence set forth in any of SEQ ID NOs: 305-309, unmodified MDA5 has the sequence set forth in SEQ ID NO: 310, unmodified RIG-I has the sequence set forth in SEQ ID NO: 311, and unmodified IRF-3 has the sequence set forth in SEQ ID NO: 312. In other examples, the therapeutic product is selected from among STING, MDA5, IRF-3, and RIG-I, and comprises a gam-of-function mutation(s) that renders the STING, MDA5, IRF-3, IRF-5, IRF-8, or RIG-I constitutively active, whereby expression of type I IFN is constitutive. Mutations can be selected as follows: a) in STING, with reference to SEQ ID NOs: 305-309, one or more selected from among: S102P, V147L, V147M, N154S, V155M, G166E, C206Y, G207E, S102P/F279L, F279L, R281Q, R284G, R284S, R284M, R284K, R284T, R197A, D205A, R310A, R293A, T294A, E296A, R197A/D205A, S272A/Q273A, R310A/E316A, E316A, E316N, E316Q, S272A, R293A/T294A/E296A, D231A, R232A, K236A, Q273A, S3 8A/E360A/S366A, D231A/R232A/K236A/R238A, S358 A, E360A, S366A, R238A, R375A, N154S/R284G, and S324A/S326A; b) in MDA5, with reference to SEQ ID NO:310, one or more of: T33 II, T331R, A489T, R822Q, G821S, A946T, R337G, D393V, G495R, R720Q, R779H, R779C, L372F, and A452T; c) in RIG-I, with reference to SEQ ID NO:311, one or both of E373A and C268F; and d) in IRF-3, with reference to SEQ ID NO:312, S396D; and e) conservative replacements of any of the above. For example, where the therapeutic product is a variant STING protein that contains one or more amino replacement(s) selected, with reference to SEQ ID NOs: 305-309, from among: S102P, V147L, V147M, N154S, V155M, G166E, C206Y, G207E, S102P/F279L, F279L, R281Q, R284G, R284S, R284M, R284K, R284T, R197A, D205A, R310A, R293A, T294A, E296A, R197A/D205A, S272A/Q273A, R310A/E316A, E316A, E316N, E316Q, S272A, R293A/T294A/E296A, D231A, R232A, K236A, Q273A, S358A/E360A/S366A, D231A/R232A/K236A/R238A, S358A, E360A, S366A, R238A, R375A, N154S/R284G, and S324A/S326A, and conservative replacements thereof.

Encoded therapeutic products, include, for example, antibodies, of any form known to those of skill in the art, and includes multi-specific, such as bi-specific antibodies, such as, for example, where the bi-specific antibody is a bi-specific T-cell engager, such as where a plasmid in the bacterium encodes a bi-specific T-cell engager antibody that binds DLL3 and CD3, such as a bi-specific T-cell engager antibody that comprises a heavy chain and light chain of an anti-DLL3 antibody and of an anti-CD3 antibody, such has those encoded in the constructs designated SC16.15, SC16.34, and SC16.56, which encode variable heavy and variable light chains of antibodies that bind each of DLL3 and CD3, and whose sequences are set forth in SEQ ID NOs.485-491, or humanized vanants thereof, and variants that have at least 95% or 98% sequence identity thereto. For example, the encoded bi-specific T-cell engager antibody comprises combinations of a) - f), whereby the resulting construct can bind to each of DLL3 and CD3: a) a light chain that comprises amino acid residues 154-260 of SEQ ID NO: 487, or a humanized variant thereof, or a variant having at least 95 % sequence identity thereto; and b) a heavy chain that comprises the sequence of amino acid residues set forth as amino acid residues 22-138 of SEQ ID NO: 487, or a humanized variant thereof, or a variant having at least 95 % sequence identity thereto; and c) a light chain that comprises a sequence of amino acid residues set forth as amino acid residues 155-261 of SEQ ID NO: 489, or a humanized variant thereof, or a variant having at least 95 % sequence identity thereto; and d) a heavy chain that comprises a sequence of amino acid residues set forth as amino acid residues 22-139 of SEQ ID NO: 489, or a humanized variant thereof, or a variant having at least 95 % sequence identity thereto; and e) a heavy and light chain, wherein: the light chain comprises a sequence of amino acid residues set forth as amino acid residues 155-261 of SEQ ID NO: 485, or a humanized variant thereof, or a variant having at least 95 % sequence identity thereto; and the heavy chain comprises a sequence of amino acid residues set forth as amino acid residues 22-139 of SEQ ID NO: 485, or a humanized variant thereof, or a variant having at least 95 % sequence identity thereto; and f) a heavy and light chain of an anti-CD3 antibody, wherein: the light chain of the anti-CD3 antibody comprises a sequence of amino acid residues set forth as amino acid residues 398-504 of SEQ ID NO: 485, or a humanized vanant thereof, or a variant having at least 95 % sequence identity thereto, or a humanized variant thereof, or a variant having at least 95 % sequence identity thereto; and the heavy chain of the anti-CD3 antibody comprises a sequence of ammo acid residues set forth as amino acid residues 267-382 of SEQ ID NO: 485, or a humanized variant thereof, or a variant having at least 95 % sequence identity thereto. The encoded bi-specific T-cell engager antibody construct can comprise a leader sequence, such as, for example, an IgGK leader sequence. The bispecific antibody also can comprise a Gly-Ser linker linking one or more light and heavy chains; and also can comprise a linker, such as a Gly-Ser linker, linking the portions that bind different targets, such as a Gly-Ser linker linking the anti-DLL3 and anti-CD3 portions of an exemplary bi- specific T-cell engager antibody. An exemplary linker comprises the sequence of amino acids set forth as residues 383-397 of SEQ ID NO: 485 and variants thereof. The bi- specific T-cell engager antibody can comprise a flag tag, such as, for example, a flag tag that comprises the sequence of amino acids set forth as residues 505-512 of SEQ ID NO: 485. Exemplary of constructs that encode a bi-specific T-cell engager antibody is a nucleic acid construct encoding a leader sequence, the heavy and light chain of an antiDLL antibody, and the heavy and light chain of an anti-CD3 antibody, and optionally one or more peptide linkers, and optionally a flag tag, such as, for example, where the encoded bi-specific T-cell engager antibody construct comprises the sequence of amino acid residues set forth in any of SEQ ID NOs: 485-491, or a humanized variant thereof, or a variant having at least 95 % sequence identity thereto, and sequences having at least 95% or 98% sequence identity thereto, and retaining the bi-specific binding.

Another example of a therapeutic product that can be encoded in the plasmid of the immunostimulatory bacteria provided is/are tumor-associated antigen(s). These plasmids can encode another therapeutic product, such as a protein that is part of a cytosolic DNA/RNA sensor pathway that leads to expression of type I interferon (IFN) part of a cytosolic DNA/RNA sensor pathway that leads to expression of type I interferon (IFN), such as, for example, a modified STING protein that constitutively induces type I interferon. Exemplary of such STING proteins is a modified STING protein that comprises the replacements corresponding to N154S, or R284G, or N154S/R284G. These include human STING, non-human STING that has a lower NF-KB signaling activity compared to human STING, and the chimeric STING proteins that comprise a CTT from a non-human STING with lower NF -KB activity than human STING. The plasmids also can encode a cytokine, that, for example, has anti-tumor or anti-viral activity, such as IL- 15 or an IL-15/IL-15R alpha chain complex. The plasmid can encode the combination of a modified STING, IL-15/IL-15R alpha chain complex, or the modified STING, IL-15, and a tumor-associated antigen and/or a bi-specific T-cell engager antibody. Other encoded therapeutic products include an immunostimulatory protein(s) that confers or contributes to an anti-tumor immune response in the tumor microenvironment that is selected from among one or more of: IL-2, IL-7, IL-12p70 (IL-12p40 + IL-12p35), IL-15, IL-2 that has attenuated binding to IL-2Ra, IL-15/IL-l 5R alpha chain complex (IL-15Ra- IL-I5sc), IL-18, IL-21, IL-23, IL-36/, IL-2 that is modified so that it does not bind to IL- 2Ra, CXCL9, CXCL10, CXCL11, interferon-a, interferon-0, interferon-y, CCL3, CCL4, CCL5, proteins that are involved in or that effect or potentiate recruitment and/or persistence of T cells, CD40, CD40 ligand (CD40L), CD28, 0X40, 0X40 ligand (OX40L), 4-1BB, 4-1BB ligand (4-1BBL), members of the B7-CD28 family, CD47 antagonists, an anti-IL-6 antibody or IL-6 binding decoy receptor, TGF-beta polypeptide antagonists, and members of the tumor necrosis factor receptor (TNFR) superfamily. Other immunostimulatory proteins that can be encoded are co-stimulatory molecules selected from among CD40, CD40 ligand (CD40L), CD28, 0X40, 0X40 ligand (OX40L), 4-1BB, and a 4-1BB ligand (4-1BBL) that optionally is truncated and lacking a cytoplasmic domain for expression on an antigen-presenting cell (APC); and where the truncated gene product is capable of constitutive immunostimulatory signaling to a T-cell through co-stimulatory receptor engagement and is unable to counter-regulatory signal to the antigen-presenting cell (APC) due to a deleted cytoplasmic domain. Other immunostimulatory proteins that confer or contribute to an anti-tumor immune response in the tumor microenvironment is an immunostimulatory protein that confers or contributes to an anti-tumor immune response in the tumor microenvironment is a cytokine, a chemokine, and/or a co-stimulatory molecule, such as cytoplasmic domain- deleted form thereof, such as one or more of 4-1BBL, CD80, CD86, CD27L, CD24L, B7RP1, and OX40L. Other therapeutic products include, for example, TGF-beta polypeptide antagonists. Other therapeutic products are antibodies or antibody or antigenbinding fragments or forms thereof, such as, but not limited to, a Fab, Fab', F(ab’)2, single-chain Fv (scFv), Fv, dsFv, nanobody, diabody fragment, and a single-chain antibody. The antibody or antigen-binding fragment thereof can be humanized or human. Exemplary of antibody and antigen binding fragments is an antagonist of PD-1, PD-L1, CTLA-4, VEGF, VEGFR2, CD24, or IL-6. The immunostimulatory bacterium can contain a plasmid that encodes two or more therapeutic products selected from among: a) an immunostimulatory protein that confers or contnbutes to an anti-tumor immune response in the tumor microenvironment; b) one or more of a protein that is part of a cytosolic DNA/RNA sensor pathway that leads to expression of type I interferon (IFN), or a variant thereof that has increased activity to increase expression of type I IFN, or a variant thereof that results in constitutive expression of a type I IFN; and c) an anti-cancer antibody or antigen-binding portion thereof. In some embodiments, the immunostimulatory protein is a co-stimulatory molecule that lacks a cytoplasmic domain or a sufficient portion thereof, for expression on an antigen-presenting cell (APC), whereby the truncated co-stimulatory molecule is capable of constitutive immunostimulatory signaling to a T-cell through co-stimulatory receptor engagement and is unable to counter-regulatory signal to the antigen presenting cell (APC). In some embodiments, the immunostimulatory bacteria comprise plasmid that encodes two or more therapeutic products under control of a single promoter, where the therapeutic products are selected from among: a) an immunostimulatory protein that confers or contributes to an anti-tumor immune response in the tumor microenvironment; b) one or more of a protein that is part of a cytosolic DNA/RNA sensor pathway that leads to expression of type I interferon (IFN), or a variant thereof that has increased activity to increase expression of type I IFN, or a variant thereof that results in constitutive expression of a type I IFN; and c) an anti-cancer antibody or antigen-binding portion thereof; and where the encoding nucleic acids are separated by an IRES sequence or 2A peptides, and each nucleic acid encoding each product is optionally operatively linked to nucleic acid encoding a signal sequence, whereby, upon translation of the encoded mRNA, each product is separately expressed and secreted from a cell comprising the bacterium and/or plasmid. Where the immunostimulatory protein is a co-stimulatory molecule, it can lack a cytoplasmic domain or a sufficient portion thereof, for expression on an antigen-presenting cell (APC), whereby the truncated co-stimulatory molecule is capable of constitutive immunostimulatory signaling to a T-cell through co-stimulatory receptor engagement and is unable to counter-regulatory signal to the antigen presenting cell (APC). For example, the immunostimulatory bacterium provided herein, wherein the plasmid encodes at least two therapeutic products selected from among a cytokine, a protein that constitutively induces a type I IFN, a co-stimulatory molecule, and an anticancer antibody or antigen-binding portion thereof. The immunostimulatory bacteria can comprise a plasmid that encodes at least two therapeutic products selected, for example, from among two or more or all of a cytokine, a protein that constitutively induces a type I IFN, a co-stimulatory molecule, and an anti-cancer antibody or antigen-binding portion thereof, and also encodes an antigen or an antigenic protein, such as one that results in an immune response against a tumor and/or a pathogen. The antigen or antigenic protein can be, for example, a tumor-associated antigen. Examples of tumor-associated antigens and proteins include, but are not limited to, an oncofetal antigen, an oncoviral antigen, and overexpressed/accumulated antigen, a cancer-Testis antigen, a linear restricted antigen, a mutated antigen, a post-translationally altered antigen, or an idiotypic antigen. Exemplary of such antigens and proteins are the following:

In some embodiments, the encoded payloads, such as the therapeutic proteins, are expressed under control of a eukaryotic promoter. In other embodiments, described herein, the encoded payloads, such as where the immunostimulatory bacterium is an RNA delivery vehicle, the payloads are expressed under control of a prokaryotic promoter recognized by the bacterium.

In all embodiments, the immunostimulatory bacteria can comprise genome modifications whereby the bacterium is flagellin", asd, msbB', pagP', and csgD'; or is ansB', asd. csgD', purl', msbB', flagellin", and pagP',' or is thyA', asd', csgD', pur!', msbB', flagellin", and pagP', or is thyA', csgD' , purl' , msbB', flagellin", and pagP', or other combinations of modifications as described herein. Included are modifications, such genome modifications of an immunostimulatory bacterium to render it less inflammatory upon administration, than wild-type. Combinations of modifications include those that render the bacterium msbBdpagP', and lacking flagella, where the wild type bacterium has flagella. The immunostimulatory bacteria provided herein can be an anti-cancer therapeutic. The immunostimulatory bacterium can be a vaccine for treating or preventing or reducing the risk of a cancer or infection from a pathogen. In such embodiments, the encoded payloads can be expressed under control of a prokaryotic promoter; and the nucleic acid encoding the payloads comprise translational regulatory signals that are recognized by eukaryotic ribosomes, and not by bacterial ribosomes. As a result, the encoded products and constructs are transcribed by the bacteria but are not translated by the bacteria; they are translated when administered to a host and when in a host cell, such as a phagocytic cell. The immunostimulatory bacteria can encode an antigen or protein or epitope(s) thereof from a pathogen. Exemplary pathogens include, but are not limited to, pathogens that cause chronic viral infections, such as infections by hepatitis viruses, herpes viruses, varicella zoster virus (VZV), Epstein-Barr virus, human immunodeficiency virus (HIV), human T-cell leukemia virus (HTLV), Respiratory Syncytial Virus (RSV), and measles virus; or is a virus or other pathogen chronically infect subjects; and pathogens that cause acute infections, such as initial infections with chronic influenza and coronaviruses, such as Severe Acute Respiratory Syndrome coronavirus (SARS-CoV), Middle East Respiratory Syndrome coronavirus (MERS-CoV), and Severe Acute Respiratory Syndrome coronavirus 2 (SARS-CoV-2, which causes COVID- 19).

In some embodiments, the plasmid encodes an antigen from a pathogen or an epitope or combination of epitopes thereof, such as an antigen from an essential viral protein, such as, for example, in the case of a coronavirus, an antigen from the Nucleocapsid, M and/or S proteins, which can result in neutralizing antibodies, and the enhancement of long-lived circulating and tissue-resident CD8+ T-cells. Provided are immunostimulatory bacteria where nucleic acid encoding the antigen, epitope, or antigenic protein is operatively linked to a prokaryotic promoter recognized by the bacterium; and the encoding sequence includes regulatory sequences for translation that are recognized by eukaryotic ribosomes, whereby the bacterium cannot translate the encoded RNA, or the encoding sequence does not include a Shine Dalgarno sequence that is recognized by the bacterial ribosomes so that the encoded mRNA is not translated; and the mRNA is delivered to the eukaryotic host cell into which the bacterium is delivered.

Any of the immunostimulatory bacteria provided herein can comprise a plasmid that encodes two or more therapeutic products under control of a single promoter, where expression of the nucleic acid encoding at least two or all of the products is under control of a single promoter, and the nucleic acid encoding each product is separated by nucleic acid resulting in separate translated products, such as nucleic acid encoding 2A polypeptides, whereby, upon translation, each product is separately expressed.

In all of the immunostimulatory bacteria provided herein, the nucleic acid encoding one or more of the therapeutic products is operatively linked to nucleic acid encoding a sequence that directs secretion of the expressed product(s).

The therapeutic products include, co-stimulatory molecule, particularly one with a cytoplasmic domain deletion for expression on an antigen-presenting cell (APC), whereby the resulting truncated gene product is capable of constitutive immunostimulatory signaling to a T-cell through co-stimulatory receptor engagement, and is unable to counter-regulatory signal to the APC due to the cytoplasmic domain deletion. Exemplary of such co-stimulatory molecules is one or more of 4-1BBL, CD80, CD86, CD27L, B7RP1, CD24L, or OX40L, particularly with the cytoplasmic domain deletion.

The immunostimulatory bacteria can encode a plurality of products. In some embodiments, at least one product is selected from a) and at least one is selected from b), wherein: a) is IL-2, IL-7, IL-12p70 (IL-12p40 + IL-12p35), IL-15, IL-23, IL-36 gamma, IL-2 that has attenuated binding to IL-2Ra, IL-15/IL-15R alpha chain complex (IL-15Ra-IL-15sc), IL- 18, IL-2 that is modified so that it does not bind to IL-2Ra, CXCL9, CXCL10, CXCL11, interferon-a, interferon-p, CCL3, CCL4, CCL5, proteins that are involved in or that effect or potentiate recruitment and/or persistence of T cells, CD40, CD40 Ligand (CD40L), 0X40, 0X40 Ligand (OX40L), 4-1BB, 4- 1BB Ligand (4-1BBL), members of the B7-CD28 family, TGF-beta polypeptide antagonists, or members of the tumor necrosis factor receptor (TNFR) superfamily; and b) is STING, RIG-I, MDA-5, IRF-3, IRF-5, IRF-7, IRF-8, TRIM56, RIP1, Sec5, TRAF3, TRAF2, TRAF6, STAT1, LGP2, DDX3, DHX9, DDX1, DDX9, DDX21, DHX15, DHX33, DHX36, DDX60, or SNRNP200.

Additional therapeutic products include, for example, one or more of a TGF- beta inhibitory antibody, a TGF-beta binding decoy receptor, an anti-IL-6 antibody, and an IL-6 binding decoy receptor. The immunostimulatory bacteria can encode one or more of the following combinations of therapeutic products:

IL-2 and IL-12p70; IL-2 and IL-21;

IL-2, IL-12p70, and a STING GOF variant;

IL-2, IL-21, and a STING GOF variant;

IL-2, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt), where Acyt is a deleted cytoplasmic domain;

IL-2, IL-21, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt);

IL-15/IL-15Ra, and a STING GOF variant;

IL-15/IL-15Ra, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt);

IL-15/IL-15Ra and IL-12p70;

IL- 15/IL- 15Ra and IL-21 ;

IL-15/IL-15Ra, IL-12p70, and a STING GOF variant;

IL-15/IL-15Ra, IL-21, and a STING GOF variant;

IL-15/IL-15Ra, IL-12p70, a STING GOF variant, and 4-1BBL (including 4- 1 BBL Acyt);

IL-15/IL-15Ra, IL-21, a STING GOF variant, and 4-1BBL (including 4-

1 BBL Acyt);

IL-12p70 and IL-21;

IL-12p70, IL-21, and a STING GOF variant;

IL-12p70, IL-21, a STING GOF variant, and 4-1BBL (including 4- 1 BBL Acyt);

IL-12p70 and a STING GOF variant;

IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt);

IL-12p70 and IL- 18;

IL-12p70, IL-18, and a STING GOF variant;

IL-12p70, IL-18, a STING GOF variant, and 4-1BBL (including 4-

1 BBL Acyt); a TGF-P decoy receptor, IL-2, and IL-12p70; a TGF-P decoy receptor, IL-2, and IL-21; a TGF- decoy receptor, IL-2, IL-12p70, and a STING GOF variant; a TGF-P decoy receptor, IL-2, IL-21, and a STING GOF variant; a TGF-P decoy receptor, IL-2, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a TGF-P decoy receptor, IL-2, IL-21, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a TGF-P decoy receptor, IL-15/IL-15Ra, and a STING GOF variant; a TGF-P decoy receptor, IL-15/IL-15Ra, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a TGF-P decoy receptor, IL-15/IL-15Ra, and IL-12p70; a TGF-P decoy receptor, IL-15/IL-15Ra, and IL-21; a TGF-p decoy receptor, IL-15/IL-15Ra, IL-12p70, and a STING GOF variant; a TGF-P decoy receptor, IL-15/IL-15Ra, IL-21, and a STING GOF variant; a TGF-P decoy receptor, IL-15/IL-15Ra, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a TGF-P decoy receptor, IL-15/IL-15Ra, IL-21, a STING GOF variant, and 4- 1BBL (including 4-lBBLAcyt); a TGF-P decoy receptor, IL-12p70, and IL-21; a TGF-p decoy receptor, IL-12p70, IL-21, and a STING GOF variant; a TGF-P decoy receptor, IL-12p70, IL-21, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a TGF-p decoy receptor and IL-12p70; a TGF-P decoy receptor, IL-12p70, and a STING GOF variant; a TGF-P decoy receptor, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a TGF-P decoy receptor, IL-12p70, and IL-18; a TGF-P decoy receptor, IL-12p70, IL-18, and a STING GOF variant; a TGF-p decoy receptor, IL-12p70, IL-18, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a TGF-P decoy receptor and a STING GOF variant; an anti-CTLA-4 antibody, IL-2, and IL-12p70; an anti-CTLA-4 antibody, IL-2, and IL-21; an anti-CTLA-4 antibody, IL-2, IL-12p70, and a STING GOF variant; an anti-CTLA-4 antibody, IL-2, IL-21, and a STING GOF variant; an anti-CTLA-4 antibody, IL-2, IL-12p70, a STING GOF variant, and 4- 1BBL (including 4-lBBLAcyt); an anti-CTLA-4 antibody, IL-2, IL-21, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); an anti-CTLA-4 antibody, IL-15/IL-15Ra, and a STING GOF variant; an anti-CTLA-4 antibody, IL-15/IL-15Ra, a STING GOF variant, and 4- 1BBL (including 4-lBBLAcyt); an anti-CTLA-4 antibody, IL-15/IL-15Ra, and IL-12p70; an anti-CTLA-4 antibody, IL-15/IL-15Ra, and IL-21; an anti-CTLA-4 antibody, IL-15/IL-15Ra, IL-12p70, and a STING GOF variant; an anti-CTLA-4 antibody, IL-15/IL-15Ra, IL-21, and a STING GOF variant; an anti-CTLA-4 antibody, IL-15/IL-15Ra, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); an anti-CTLA-4 antibody, IL-15/IL-15Ra, IL-21, a STING GOF variant, and 4-1 BBL (including 4-lBBL cyt); an anti-CTLA-4 antibody, IL-12p70, and IL-21; an anti-CTLA-4 antibody, IL-12p70, IL-21, and a STING GOF variant; an anti-CTLA-4 antibody, IL-12p70, IL-21, a STING GOF variant, and 4- 1BBL (including 4-lBBLAcyt); an anti-CTLA-4 antibody and IL-12p70; an anti-CTLA-4 antibody, IL-12p70, and a STING GOF variant; an anti-CTLA-4 antibody, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); an anti-CTLA-4 antibody, IL-12p70, and IL-18; an anti-CTLA-4 antibody, IL-12p70, IL- 18, and a STING GOF variant; an anti-CTLA-4 antibody, IL-12p70, IL- 18, a STING GOF variant, and 4- 1BBL (including 4-lBBLAcyt); an anti-CTLA-4 antibody and a STING GOF variant; a CD40 agonist, IL-2, and IL-12p70; a CD40 agonist, IL-2, and IL-21; a CD40 agonist, IL-2, IL-12p70, and a STING GOF variant; a CD40 agonist, IL-2, IL-21, and a STING GOF variant; a CD40 agonist, IL-2, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a CD40 agonist, IL-2, IL-21, a STING GOF variant, and 4-1BBL (including 4- IBBLAcyt); a CD40 agonist, IL-15/IL-15Ra, and a STING GOF variant; a CD40 agonist, IL-15/IL-15Ra, a STING GOF vanant, and 4-1BBL (including 4-lBBLAcyt); a CD40 agonist, IL-15/IL-15Ra, and IL-12p70; a CD40 agonist, IL-15/IL-15Ra, and IL-21; a CD40 agonist, IL-15/IL-15Ra, IL-12p70, and a STING GOF variant; a CD40 agonist, IL-15/IL-15Ra, IL-21, and a STING GOF variant; a CD40 agonist, IL-15/IL-15Ra, IL-12p70, a STING GOF variant, and 4- 1BBL (including 4-lBBLAcyt); a CD40 agonist, IL-15/IL-15Ra, IL-21, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a CD40 agonist, IL-12p70, and IL-21; a CD40 agonist, IL-12p70, IL-21, and a STING GOF variant; a CD40 agonist, IL-12p70, IL-21, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a CD40 agonist and IL-12p70; a CD40 agonist, IL-12p70, and a STING GOF variant; a CD40 agonist, IL-12p70, a STING GOF variant, and 4-1BBL (including 4- IBBLAcyt); a CD40 agonist, IL-12p70, and IL-18; a CD40 agonist, IL-12p70, IL-18, and a STING GOF variant; a CD40 agonist, IL-12p70, IL-18, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a tumor-associated antigen; a CD40 agonist and a STING GOF variant. STING GOF variants include chimeric STING and non-human STING, including those described or exemplified herein, including those detailed above and below.

In all embodiments, the immunostimulatory bacterium provided herein, can also encode a tumor-associated antigen. These bacteria are of interest for use as vaccines and as therapeutics. Other exemplary combinations of encoded products include:

IL-2 and IL-12p70;

IL-2 and IL-21;

IL-2, IL-12p70, and a STING GOF variant;

IL-2, IL-21, and a STING GOF variant; IL-2, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt), where Acyt is a deleted cytoplasmic domain;

IL-2, IL-21, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt);

IL-15/IL-15Ra, and a STING GOF variant;

IL-15/IL-15Ra, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt);

IL-15/IL-15Ra and IL-12p70;

IL- 15/IL- 15Ra and IL-21 ;

IL-15/IL-15Ra, IL-12p70, and a STING GOF variant;

IL-15/IL-15Ra, IL-21, and a STING GOF variant;

IL-15/IL-15Ra, IL-12p70, a STING GOF variant, and 4-1BBL (including 4- 1 BBL Acyt);

IL-15/IL-15Ra, IL-21, a STING GOF variant, and 4-1BBL (including 4-

1 BBL Acyt);

IL-12p70 and IL-21;

IL-12p70, IL-21, and a STING GOF variant;

IL-12p70, IL-21, a STING GOF variant, and 4-1BBL (including 4- 1 BBL Acyt);

IL-12p70 and a STING GOF variant;

IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt);

IL-12p70 and IL- 18;

IL-12p70, IL-18, and a STING GOF variant;

IL-12p70, IL-18, a STING GOF variant, and 4-1BBL (including 4-

1 BBL Acyt); a TGF-P decoy receptor, IL-2, and IL-12p70; a TGF-P decoy receptor, IL-2, and IL-21; a TGF- decoy receptor, IL-2, IL-12p70, and a STING GOF variant; a TGF-P decoy receptor, IL-2, IL-21, and a STING GOF variant; a TGF-P decoy receptor, IL-2, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a TGF-P decoy receptor, IL-2, IL-21, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a TGF-P decoy receptor, IL-15/IL-15Ra, and a STING GOF variant; a TGF-P decoy receptor, IL-15/IL-15Ra, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a TGF-P decoy receptor, IL-15/IL-15Ra, and IL-12p70; a TGF-P decoy receptor, IL-15/IL-15Ra, and IL-21; a TGF-P decoy receptor, IL-15/IL-15Ra, IL-12p70, and a STING GOF variant; a TGF-P decoy receptor, IL-15/IL-15Ra, IL-21, and a STING GOF variant; a TGF-P decoy receptor, IL-15/IL-15Ra, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a TGF-P decoy receptor, IL-15/IL-15Ra, IL-21, a STING GOF variant, and 4- 1BBL (including 4-lBBLAcyt); a TGF-P decoy receptor, IL-12p70, and IL-21; a TGF-P decoy receptor, IL-12p70, IL-21, and a STING GOF variant; a TGF-P decoy receptor, IL-12p70, IL-21, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a TGF-p decoy receptor and IL-12p70; a TGF-P decoy receptor, IL-12p70, and a STING GOF variant; a TGF-P decoy receptor, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a TGF-P decoy receptor, IL-12p70, and IL-18; a TGF-P decoy receptor, IL-12p70, IL-18, and a STING GOF variant; a TGF-p decoy receptor, IL-12p70, IL-18, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a TGF-P decoy receptor and a STING GOF variant; an anti-CTLA-4 antibody, IL-2, and IL-12p70; an anti-CTLA-4 antibody, IL-2, and IL-21; an anti-CTLA-4 antibody, IL-2, IL-12p70, and a STING GOF variant; an anti-CTLA-4 antibody, IL-2, IL-21, and a STING GOF variant; an anti-CTLA-4 antibody, IL-2, IL-12p70, a STING GOF variant, and 4- 1BBL (including 4-lBBLAcyt); an anti-CTLA-4 antibody, IL-2, IL-21, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); an anti-CTLA-4 antibody, IL-15/IL-15Ra, and a STING GOF variant; an anti-CTLA-4 antibody, IL-15/IL-15Ra, a STING GOF variant, and 4- 1BBL (including 4-lBBLAcyt); an anti-CTLA-4 antibody, IL-15/IL-15Ra, and IL-12p70; an anti-CTLA-4 antibody, IL-15/IL-15Ra, and IL-21; an anti-CTLA-4 antibody, IL-15/IL-15Ra, IL-12p70, and a STING GOF variant; an anti-CTLA-4 antibody, IL-15/IL-15Ra, IL-21, and a STING GOF variant; an anti-CTLA-4 antibody, IL-15/IL-15Ra, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); an anti-CTLA-4 antibody, IL-15/IL-15Ra, IL-21, a STING GOF variant, and 4-1 BBL (including 4-lBBLAcyt); an anti-CTLA-4 antibody, IL-12p70, and IL-21; an anti-CTLA-4 antibody, IL-12p70, IL-21, and a STING GOF variant; an anti-CTLA-4 antibody, IL-12p70, IL-21, a STING GOF variant, and 4- 1BBL (including 4-lBBLAcyt); an anti-CTLA-4 antibody and IL-12p70; an anti-CTLA-4 antibody, IL-12p70, and a STING GOF variant; an anti-CTLA-4 antibody, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); an anti-CTLA-4 antibody, IL-12p70, and IL-18; an anti-CTLA-4 antibody, IL-12p70, IL- 18, and a STING GOF variant; an anti-CTLA-4 antibody, IL-12p70, IL- 18, a STING GOF variant, and 4- 1BBL (including 4-lBBLAcyt); an anti-CTLA-4 antibody and a STING GOF variant; a CD40 agonist, IL-2, and IL-12p70; a CD40 agonist, IL-2, and IL-21; a CD40 agonist, IL-2, IL-12p70, and a STING GOF variant; a CD40 agonist, IL-2, IL-21, and a STING GOF variant; a CD40 agonist, IL-2, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a CD40 agonist, IL-2, IL-21, a STING GOF variant, and 4-1BBL (including 4- IBBLAcyt); a CD40 agonist, IL-15/IL-15Ra, and a STING GOF variant; a CD40 agonist, IL-15/IL-15Ra, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a CD40 agonist, IL-15/IL-15Ra, and IL-12p70; a CD40 agonist, IL-15/IL-15Ra, and IL-21; a CD40 agonist, IL-15/IL-15Ra, IL-12p70, and a STING GOF vanant; a CD40 agonist, IL-15/IL-15Ra, IL-21, and a STING GOF variant; a CD40 agonist, IL-15/IL-15Ra, IL-12p70, a STING GOF variant, and 4- 1BBL (including 4-lBBLAcyt); a CD40 agonist, IL-15/IL-15Ra, IL-21, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a CD40 agonist, IL-12p70, and IL-21; a CD40 agonist, IL-12p70, IL-21, and a STING GOF variant; a CD40 agonist, IL-12p70, IL-21, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a CD40 agonist and IL-12p70;** a CD40 agonist, IL-12p70, and a STING GOF variant; a CD40 agonist, IL-12p70, a STING GOF variant, and 4-1BBL (including 4- IBBLAcyt); a CD40 agonist, IL-12p70, and IL-18; a CD40 agonist, IL-12p70, IL-18, and a STING GOF variant; a CD40 agonist, IL-12p70, IL-18, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a CD40 agonist and a STING GOF variant; a bi-specific T-cell engager (BiTe) + a STING protein, a BiTe + IL-15, a BiTe + IL-15 + a STING protein, where the BiTe targets DLL3, EGFR, Her2, CEA, Mesothelin, PSMA, EpCAM, CD74, Folate Receptor, Nectin4, EphA2, CA-IX, B7H3, Siglec-15, Mucl, or Lewis Y antigen; a tumor antigen(s) + STING gain-of-function variant; a therapeutic composition of a tumor antigen(s) and IL- 15; a therapeutic composition of a tumor antigen(s) + IL- 15 + a STING gain-of- function variant; one or more antigens and an IFN; one or more antigens and an IFNot; one or more antigens, and IFNa2 or an IFNal-16; one or more antigens and any of IFNal-16; one or more antigens and IFN-P; one or more antigens, IFNa2, and IFN-P; one or more antigens and an IRF3 GOF variant with the mutation S396D; one or more antigens, IFNa2 or an IFNal-16, and an IRF3 GOF variant with the mutation S396D;

IFNalpha2 + IRF3-S396D;

IFNal-16 + IRF3-S396D;

IFNalpha2 + IFN-beta;

IFNal-16 + IFN-beta

FLT-3L, or sialidase, or IL-12p35, or Azurin, or a membrane anchored IL-2, IL-12, IL-12p35, IL-21, IL-15, FLT-3L, alone or in combination with other immunostimulatory proteins; and a TLR8 agonist, where the agonist is polyU or polyU/G, a microRNA, or miR- 21, alone or in combination with any of the immunostimulatory proteins. In all embodiments, the immunostimulatory bacteria can encode a tumor-associated antigen, such as, for example, any listed in the table above, and described herein.

Other encoded therapeutic products include, for example, a bi-specific T-cell engager, such as, for example, one that binds Delta-like ligand 3 (DLL3) and CD3. In all embodiments, the immunostimulatory bacteria can encode a cytokine, such as a cytokine, and a modified or variant STING protein. The immunostimulatory bacteria a encode immunostimulatory protein(s) that confers or contributes to anti-tumor immunity in the tumor microenvironment, such as a cytokine or a chemokine that confer or contribute to anti-tumor immunity in the tumor microenvironment. Exemplary of cytokines are IL- 15, IL-2, and IL- 12, such as IL-15/IL-15R alpha chain complex. Exemplary of cytokines are IL-15, IL-2, and IL-12, such as IL-15/IL-15R alpha chain complex. Exemplary of STING proteins are any described herein, particularly those that include gain-of-function mutations so that the STING protein constitutively induces type I IFN. The STING protein also can be modified or be selected so that it has lower NF-KB signaling activity than human STING, so that NF -KB signaling is low, and type I IFN induction is constitutive.

Exemplary STING proteins is a chimeric STING protein that comprises a human STING protein with the CTT from Tasmanian devil, or is chimeric STING that comprises a human STING protein with the CTT from Tasmanian devil and having one or more gain-of-function mutations, such as one or both of N154S and R284G or any of the mutations described herein or known in the art to effect constitutive activity. Exemplary modified STING gain-of-function variant are any described herein.

The encoded therapeutic products can comprise a multimerization domain, such as an Fc domain. Other encoded therapeutic products include any described herein, such as a product that is a B7 protein transmembrane domain, and/or a bi- specific T-cell engager antibody. The encoded therapeutic products can be GPI- anchored, or include moi eties, such as polypeptides, such as human serum albumin (HSA) or a portion thereof, that increase serum half-life of the encoded product. The therapeutic products can be other fusion proteins, such as a fusion to collagen.

The immunostimulatory bacterium can be derived from any suitable bacterial species, including, but are not limited to, species such as Salmonella, Listeria, andE. coli, and any listed or described herein. The immunostimulatory bacterium contain genome modifications, whereby the bacteria do not infect or have reduced infectivity of epithelial cells, and modified LPS to attenuate the bacteria and/or to increase uptake or infection of phagocytic cells, such as tumor-resident macrophage, and to increase tumor colonization. Various genome modifications are described herein that effect such properties. Strains include those that lack flagella and have penta-acylated LPS. Exemplary strains, include those designated YS1646Aasd/AFLG/ApagP/AansB/AcsgD/F-ApurI, or YS1646Aasd/AFLG/ApagP/AansB/AcsgD/F-ApurEAthyA, and other strains that contain genome modifications whereby the bacteria are adenosine auxotrophs, lack flagella, have penta-acylated LPS, such as by genome modifications that render the bacteria msbRIpagP', and optionally lack or have a reduction in curli fimbriae.

The bacteria provided herein are useful as therapeutics for diseases, disorders, and conditions, such as cancers. They also are useful as vaccines to prevent (reduce the risk of the diseases, disorders, and conditions, or the severity thereof) or treat diseases, disorders, and conditions, such as one caused by a pathogen or cancers. They can be designed, as described herein, so that they deliver RNA. Provided are genome modified bacteria that comprise genome modifications, whereby the response by toll-like receptors (TLRs) 2, 4, and 5 is reduced compared to the bacterium without the genome modifications, wherein: the bacterium comprises further genomic modifications whereby it is auxotrophic for a required nutrient or factor so that it is unable to replicate in a eukaryotic host, but can replicate in vitro when supplied with the nutrient or factor; the bacterium comprises a plasmid containing nucleic acid encoding a product, or comprises RNA encoding the product; the product encoded by the nucleic acid or RNA is an antigenic sequence or sequences from pathogen that is a pathogenic virus, bacterium, or parasite, or is a tumor antigen, whereby, upon expression of the encoded antigen in the host, the host develops an immune-protective response or immunizing response against the pathogenic virus, bacterium, parasite, or tumor antigen, or the product is a therapeutic product; expression of the antigenic sequence(s) is/are under control of a prokaryotic promoter so that RNA encoding the antigen(s) is produced in the bacterium; nucleic acid encoding the antigen comprises regulatory sequences that inhibit or prevent translation of encoded RNA by bacterial ribosomes, but that does not inhibit or prevent translation of the encoded RNA by eukaryotic host ribosomes, whereby translation is de-coupled from transcription in the bacterium; the resulting bacterium is selective for infecting phagocytic cells when administered to a eukaryotic subject, and delivers the nucleic acid into the phagocytic cells, wherein the RNA is translated.

These immunostimulatory bacterium can encode a plurality of products, which can be encoded as polycistronic message or under control of separate promoters, or in any suitable configuration. The nucleic acid encoding the products or at least the antigenic sequence(s) can comprise sequences that prevent or inhibit translation by a prokaryotic host, such as the bacterium, and/or include sequences that facilitate translation in a eukaryotic host, while inhibiting or preventing translation in the bacterium. Exemplary of such sequences is an internal ribosomal entry site (IRES) sequence, whereby host cell translation is facilitated or enhanced, and bacterial translation is inhibited or prevented. As a result the products encoded in the bacterium are transcribed into RNA, but not translated until they are in a eukaryotic host. The bacteria thereby serve as RNA delivery vehicles. Exemplary of an IRES is where the IRES is a Vascular Endothelial Growth Factor and Type 1 Collagen Inducible Protein (VCIP) IRES. Exemplary bacterium are provided, where the nucleic acid encoding the antigen(s) comprises a VCIP or other IRES that inhibits or reduces translation in the bacterium, and permits and optionally promotes or enhances translation in a eukaryotic host. The translational regulatory sequence, such as the IRES or the VCIP IRES, can be included in the plasmid, at a position that is 3’ of the promoter and 5’ of the antigen(s) coding sequence. An exemplary of a VCIP IRES is set forth in SEQ ID NO:434 or is a sequence that has at least 98% sequence identity therewith and has activity as an IRES.

Pathogens from which or against which the encoded antigens are derived include any pathogen, such as a bacterium or a virus. The encoded antigens also include tumor antigens. The resulting bacterium can be a vaccine to prevent or treat a viral infection or a bacterial infection or to prevent or treat a cancer. The pathogen can be selected from among viruses that causes chronic viral infections. Exemplary of infections are infections caused by hepatitis viruses, herpes viruses, varicella zoster virus (VZV), Epstein-Barr virus, human immunodeficiency virus (HIV), human T-cell leukemia virus (HTLV), Respiratory Syncytial Virus (RSV), measles virus, and other viruses that chronically infect subjects. The infection can be an acute infection, such as an infection caused by Severe Acute Respiratory Syndrome coronavirus (SARS- CoV), Middle East Respiratory Syndrome coronavirus (MERS-CoV), or Severe Acute Respiratory Syndrome species of Escherichia, Staphylococcus, Pseudomonas, Actinobacteria, Archaeobacteria, Mycobacteria or Porphyromonas. Other pathogens include / ³ , gingivalis, SARS-CoV2, or E. coli, or Haemophilus influenza.

The plasmid in the bacterium can encode an antigen, and also can further encode an immunostimulatory protein or other adjuvant, as well as a combination of immunostimulatory proteins or other therapeutic proteins, such as a STING protein particularly a modified STING that comprises gain-of-function mutation and/or a chimeric STING protein, such as any described herein. Hence any of the bacterium described herein can be provided so that they deliver mRNA encoding any of the products and combinations described herein. The products can be encoded in the plasmid as part of a polycistronic sequence with expression of the antigen under control of a prokaryotic promoter recognized by the bacterium; or the immunostimulatory protein(s) and/or other therapeutic proteins are encoded on the plasmid under control of a eukaryotic promoter recognized by the eukaryotic host. The resulting bacterium can comprise mRNA encoding the antigen(s) and any other proteins expressed under control of a prokaryotic promoter, where the mRNA is produced by culturing the bacterium in vitro. The bacteria include those that comprises genome modifications whereby the bacterium lacks flagella and produces LPS with penta-acylated lipid A. The bacteria can optionally be asd or thyA' or both, and/or one or both of an adenosine auxotroph, and csgD', and is optionally ansB'. The bacterium can comprise or further comprise nucleic acid encoding a TLR8 agonist, such as, for example, polyU, polyU/G, a microRNA, or miR-21. Exemplary of the bacteria is one that is msbB'/pagP~, lacks flagella, and is asd or thyA~ or both asd and thyA~. Exemplary bacterial species, include, but are not limited to, a species or strain of Escherichia, Listeria, Mycobacteria, or Salmonella. The bacterium can be a strain of Salmonella, such as a Salmonella typhimurium. The unmodified Salmonella can be a wild-type strain, or the unmodified Salmonella strain can be an attenuated strain. Exemplary of the starting bacteria are those derived from strain VNP20009 or YS1646, or from strain ATCC 14028, or from a strain having all of the identifying characteristics of strain ATCC 14028.

Genome modifications include any modification that results in a change the nucleic acid sequence, and generally a phenotype. Genome modifications include one or more of a deletion, insertion, disruption, transposition, and other modification in a gene, whereby the product encoded by the gene is not produced or, as produced, is inactive.

Promoters that control expression of the encoded products in the plasmid can be prokaryotic promoters, particularly in embodiments in which the bacterium is provided as a vaccine and/or as an RNA delivery vehicle. These include embodiments in which the product is expressed in vitro in the bacterium prior to administration to a subject, such as a human or animal. Prokaryotic promoters include bacterial promoters and bacterial phage promoters. Any promoter recognized by the bacterial RNA polymerase or by an encoded phage polymerase.

Provided are vaccines, comprising ay of the bacteria provided herein, particularly those that encode antigens for immunization for treatment or prevention, in an amount and in a vehicle for administration into a subject to elicit an adaptive immune response in a subject. The vaccine and other compositions containing the bacteria provided herein can be formulated for any route of administration, such as formulated as an aerosol, or as a powder, or as a tablet, or a suppository. They can be formulated for oral administration, nasal administration, inhalation administration, rectal administration, vaginal administration, intraocular administration, intracranial administration, intradermal administration, or intramuscular administration.

Provided are vaccines that comprise nucleic acid encoding an antigen from a protein from a viral pathogen, such as a respiratory virus, such as a corona virus, such as SARS-COV2, formulated for nasal or pulmonary inhalation. The vaccine is formed and formulated such that it does not sufficiently activate TLR2, whereby the vaccine induces type I IFN. Activation of TLR2 inhibits or reduces activation of type I IFN; thus, a vaccine that does not activate TLR2 or does so at a low enough level so that type I IFN is activated, such as by the immunostimulatory bacteria and vaccines provided herein.

The vaccines, also are formed/formulated so that they do not activate or have low enough activation of a TLR4 and/or TLR5 response sufficient to decease or inhibit type I IFN, so that type I IFN is expressed. As described herein, many vaccines and delivery vectors designed to stimulate type I IFN expression, also have properties that activate TLR2, 4 and/or 5, which activation is at level sufficient to inhibit or reduce type I IFN expression. The immunostimulatory bacterium and vaccines provided herein are designed so that they do not sufficiently activate particularly TLR2, and also TLR4 and/or TLR5, so that expression of type I IFN is not reduced or inhibited by the bacterium or vaccine.

Provided are vaccines that comprise nucleic acid that encodes an antigen or protein or epitope from a pathogen or tumor, where the vaccine elicits an immune response against the pathogen or tumor; the pathogen is a respiratory pathogen that infects the respiratory system including the lungs and/or naso-pharynx; the tumor is a lung or respiratory tract tumor; the vaccine is formulated for inhalation through the nose or lungs; the vaccine delivers the nucleic acid to phagocytic macrophages to convert the immunosuppressive phagocytic macrophages to immunostimulatory, phagocytic macrophages that are capable of in situ antigen cross-presentation to CD8+ T-cells, and of migration to lymph nodes to prime CD4+ and CD8+ T-cells. The vaccine is designed, formed, and/or formulated so that it does not activate a TLR4 and/or TLR5 response sufficient to decease or inhibit type I IFN. Vaccines that do not activate TLR2/4/5 sufficiently to decrease or inhibit type IF are provided. The vaccines can encode product so that the vaccine elicits an immune response against a pathogen, such as a virus. Virus pathogens include mRNA viruses, such as a corona virus or influenza virus. Corona viruses include, SARS virus, such as a SARS-COV2 virus. The vaccines encode an antigen, protein, or epitope of from a pathogen, such as a viral antigen, protein, or epitope. Exemplary of proteins is a capsid or nucleoprotein. For example, where the virus is SARS-COV2, the protein or epitope is or is from a protein designated or encoded by SI, S2, Envelope (E), Membrane (M), Nucleocapsid (N), ORF3a, ORF6, ORF7a, ORF7b, and ORF8, such as a protein or epitope that is or is from or is a spike protein. The vaccines include the bacteria provided herein that deliver mRNA, such as mRNA encoding the protein or antigen. The mRNA include mRNA that is modified to increase stability of the mRNA and/or the stability of the encoded protein or antigen or epitope. The encoded proteins can be modified, such as modifications that alter the structure of the protein to alter interaction with host cell proteins. For example, mRNA and proteins have been designed that improve or increase or stabilize the interaction of the encoded protein or epitope with a cell surface receptor. Modified mRNA encoding the spike protein from a SARS-COV2 virus has been designed; the skilled person similarly can design other modified proteins/mRNA from other viruses to enhance or improve the efficacy in preventing or reducing or ameliorating the disease caused by the virus. The vaccines can be delivery vehicles, such as oncolytic viruses and immunostimulatory bacteria, that comprise nucleic acid that encodes an antigen or protein from a pathogen, or encodes a tumor antigen.

The immunostimulatory bacteria comprise genome modifications so that they have penta-acyl ated lipopolysaccharides (LPS), and lacks flagella, wherein wild-type bacterium has flagella. As a result, the bacteria do not elicit an inflammatory response or elicit a reduced inflammatory response compared, for example, to the bacterium designated VNP20009. Additionally, the bacteria can comprise genome modifications whereby they do not produce curli fimbriae. In addition to encoding immunostimulatory proteins, such as a cytosolic DNA RNA sensor pathway proteins, such as eSTING and a cytokine, such as an IL-15/IL-15R alpha chain complex or IL- 15, they can encode a tumor-associated antigen (TAA). Products that are part of a cytosolic DNA/RNA sensor pathway that leads to expression of type I interferon (IFN), include for example, STING, IRF3, IRF5, IRF7, IRF8, MDA5, RIG-I, and particularly modified forms thereof that comprises a gain-of-function mutation, whereby expression of the type I interferon is constitutive. The immunostimulatory bacteria can be an attenuated bacterium or a Gram-negative bacterium, or is a Grampositive bacterium.

Exemplary bacteria from which the immunostimulatory bacteria can be derived, include, but are not limited to strains of Salmonella, Shigella, E. coli, Bifidobacteriae, Rickettsia, Vibrio, Listeria, Klebsiella, Bordetella, Neisseria, Aeromonas, Francisella, Cholera, Corynebacterium, Citrobacter, Chlamydia, Haemophilus, Brucella, Mycobacterium, Mycoplasma, Legionella, Rhodococcus, Pseudomonas, Helicobacter, Bacillus, or Erysipelothrix, or Archaeobacteria, an attenuated strain thereof or a modified strain thereof of any of the preceding list of bacterial strains. The bacterium can be a strain, for example, of Shigella, E. coli, Listeria, or Salmonella.

For example, the bacterium can be a Rickettsia rickettsiae, Rickettsia prowazekii, Rickettsia tsutsugamuchi, Rickettsia mooseri, Rickettsia sibirica, Bordetella bronchiseptica, Neisseria meningitidis, Neisseria gonorrhoeae, Aeromonas eucrenophila, Aeromonas salmonicida, Francisella tularensis, Corynebacterium pseudotuberculosis, Citrobacter freundii, Chlamydia pneumoniae, Haemophilus somnus, Brucella abortus, Mycobacterium intracellular e, Mycobacterium tuberculosis, Staphylococcus aureus, Legionella pneumophila, Rhodococcus equi, Pseudomonas aeruginosa, Helicobacter mustelae, Vibrio cholerae, Bacillus subtilis, Erysipelothrix rhusiopathiae, Yersinia enterocolitica, Rochalimaea quintana, or Agrobacterium tumerfacium bacterium. Included are Salmonella typhimurium strains, such as an unmodified Salmonella is a wild-type strain, such as, for example, the unmodified Salmonella strain is attenuated, or for example, where the immunostimulatory bacterium is derived from strain VNP20009 or YS1646, or strain ATCC 14028, or a strain having all of the identifying characteristics of strain ATCC 14028.

The immunostimulatory bacteria can be ansB', asd, csgD' ,purl' , msbB', flagellin', and pagP' or that is ansB', thyA', csgD', purl', msbB', flagellin', and pagP'.

The bacterium can be modified to encode and express the gene resistance to complement killing (rck), such as, for example, Salmonella rck gene, such as an E. coli strain, such as Nissle, that is modified to express rck.

Also provided are pharmaceutical compositions that contain the therapeutics provided herein, including any of the immunostimulatory bacteria in a pharmaceutically acceptable vehicle. They can be formulated for systemic administration, such as formulated for parenteral administration, or intravenous administration, or intramuscular administration, or intratumoral administration, or intraperitoneal administration, or oral administration, or rectal administration, or vaginal administration, or intraocular administration, or intradermal administration, or intracranial administration, or mucosal administration, or oral, or administration by inhalation into the mouth or nose, or, for example, by rectal, or by aerosol into the lung and/or nose, or mucosal, or intracranial, or intradermal, or intratumoral. Methods of treating cancer and uses of the therapeutics, including the immunostimulatory bacteria are provided. The cancers can comprise a solid tumor or a hematological malignancy or any other malignancy The methods comprise administering the compositions. The subjects can be selected by, for example a biopsy to identify subjects whose tumors comprise proliferating macrophages, such as proliferating M2 macrophage by identifying macrophages with markers of proliferation, such as biopsy surface markers: CD68 + KI67 and/or PCNA, MERTK. Proliferating macrophage can exhibit all of the above markers or a subset thereof. For example, gene expression of the G2M module, where more than half (>14 genes of the set) are expressed. Additionally STMN1 + the G2M module can be used to confirm proliferating. Alternatively tumor macrophage can be biopsied and assed for expression of at least two of CD68, MERTK, and K167 and/or PCNA.

Combination therapies also are provided, such as regiments in which the subject is first treated with an agent, such as a chemotherapeutic that induces apoptosis in tumors, or a checkpoint inhibitor, such as an anti-PD-1 or anti-PD-Ll antibody, prior to administration of therapeutics provided herein. The second anticancer agent or treatment is administered before, concomitantly with, after, or intermittently with, the immunostimulatory bacterium, or pharmaceutical composition. The second anti-cancer agent or treatment can be an immunotherapy, or a chemotherapy, or surgery, or radiation, or combinations thereof.

Treated cancers include, but are not limited to, a cancer is selected from among leukemia; lymphoma; gastric cancer; and cancer of the breast, heart, lung, small intestine, colon, spleen, kidney, bladder, head and neck, colorectum, ovary, prostate, brain, pancreas, skin, bone, bone marrow, blood, thymus, uterus, testicles, cervix, and liver. The cancer can be metastatic.

Second agents include, but are not limited to, agents selected from among an anti-PD-1, anti-PD-Ll, or anti-CTLA-4 antibody, anti-IL-6, anti-Siglec-15, anti- VEGF, anti-CD73, and anti-CD38 antibodies. Other exemplary second agents, can be selected from among a poly (ADP-ribose) polymerase (PARP) inhibitor, a histone deacetylase (HDAC) inhibitor, a chemotherapy agent, an anti-EGFR antibody, a CAR-T cell, an anti-Her2 antibody, an anti-mesothelin antibody, and an anti-B-cell maturation antigen (BCMA) antibody.

Pharmaceutical Compositions Provided are pharmaceutical composition, comprising any of the bacteria and/or vaccines provided herein. The bacterium or other delivery vehicle is formulated in in a pharmaceutically acceptable vehicle. The formulation can be a liquid, powder, such as a lyophilized powder, or tablet, or other suitable formulation. The vaccines, in particular, can be locally administered, such as, but not limited to, by inhalation, intramuscular, and transdermal administration, so that a local immune response can result to prevent or reduce the likelihood of or severity of an infection. For treating tumors, the bacteria and vaccines can be administered systemically, such as by intravenous administration, or can be ad mistered by intratumoral, or other route, such as intrahepatic, peritoneal, and other modes. The immunostimulatory bacteria accumulate and colonize phagocytic cells, particularly those at the locus of administration, and/or in phagocytic cells in the tumor microenvironment and in tumors.

Methods and Uses

The vaccines and bacteria and pharmaceutical compositions provided herein are for use for treatment of diseases, disorders, and conditions, including cancer and infections, and for prevention or treatment or reduction in symptoms thereof. Provided are bacteria, referred to herein as immunostimulatory bacteria by virtue of their ability to accumulate in and phagocytic cells, including tumor-resident macrophage, and to stimulate an immune response by virtue of their properties and composition, additionally by virtue of the encoded payloads, and the combination of the properties and structure of the bacteria and the payloads. The bacteria can be used as therapeutics for treatment of diseases, disorders, and conditions, and by virtue of the components of the nucleic acid constructs in the plasmids in the bacteria can produce encoded proteins, and also can be used to deliver mRNA.

Methods of treatment of cancer and/or viral other pathogen infections are provided. The uses and methods include administering or using the RNA delivery systems and vaccines and immunostimulatory bacteria provided herein for treating or preventing (reducing the risk of or severity of the diseases, disorders, and condition) cancer and/or a viral infection. The immunostimulatory bacteria and vaccines are those that encode a tumor-associated antigen or viral or other pathogen antigen, protein or epitope.

Hence, provided are uses and methods of treatment using the immunostimulatory bacteria, vaccines, delivery vehicles and composition provided herein for use for treating or preventing (reducing the risk of developing) a disease or condition or infection or cancer.

Also provided are methods of converting an M2 macrophage to an Ml or Ml- like phenotype, by administering the immunostimulatory bacterium that is modified as described herein, and that encodes an immunostimulatory protein, such STING, particularly a modified STING, and combinations of the modified STING with a cytokine, such as an IL- 15, such as IL-15/IL-15R alpha chain complex. These immunostimulatory bacteria are used for treating or administered to a subject with a condition, disease, or disorder treated by enhancing an anti-viral or anti-tumor immune response. Uses of the immunostimulatory bacteria to convert an M2 macrophage to an Ml or Ml -like phenotype macrophage in a subject with a condition, disease, or disorder treated by enhancing an anti-viral or anti-tumor immune response. The subject can have a disease, disorder, or condition that is a cancer and/or a viral or other pathogen infection.

Provided are methods of delivering RNA encoding a therapeutic product, comprising administering an immunostimulatory bacterium, designed as described above, and below, to deliver RNA, for treatment of a disease, condition, or disorder. Uses of such bacteria for treatment also are provided. Diseases, disorders, and conditions, include cancers and/or a viral or other pathogen infections. As described herein, the immunostimulatory bacteria are used to transcribe the encoded products in vitro, but not translate them, so that they deliver RNA, such as mRNA, when administered to a subject. Provided are bacteria for use for delivering RNA to a subject, comprising a plasmid encoding a heterologous product, where: the nucleic acid encoding the heterologous product is linked to a promoter recognized by the bacterium; and the nucleic acid encoding the product comprises eukaryotic sequences for translation that are not recognized by the bacterium, whereby the bacterium produces RNA, but does not translate the RNA. The bacteria deliver RNA encoding a therapeutic product and/or an antigen or protein from a pathogen or tumor for eliciting an immune response against the antigen or protein.

The encoded therapeutic products include any described herein and in the original claims, such as nucleic acid encoding a protein that is part of a cytosolic DNA/RNA sensor pathway that leads to expression of type I interferon (IFN), or a variant thereof. Type I IFNs include interferon-a and interferon-p. Variants include those that, when expressed in a subject, lead to constitutive expression of type I IFN. These include a gain-of-function (GOF) vanant that does not require cytosolic nucleic acids, nucleotides, dinucleotides, or cyclic dinucleotides (CDNs) to result in expression of type I IFN. Exemplary of these proteins is a protein selected from among STING, RIG-I, MDA-5, IRF-3, IRF-5, IRF-7, IRF-8, TRIM56, RIP1, Sec5, TRAF3, TRAF2, TRAF6, STAT1, LGP2, DDX3, DHX9, DDX1, DDX9, DDX21, DHX15, DHX33, DHX36, DDX60, and SNRNP200, and variants thereof that have increased activity, or that result in constitutive expression of type I interferon (IFN). Variants include a variant of STING, RIG-I, IRF-3, or MDA5, in which one or more serine (S) or threonine (T) residue(s) that is/are phosphorylated as a consequence of viral infection, is/are replaced with an aspartic acid (D) residue, whereby the resulting variant is a phosphomimetic that constitutively induces type I IFN, and any known to those of skill in the art and/or described herein. Variants include, for example, those wherein the mutations are selected as follows: a) in STING, with reference to SEQ ID NOs: 305-309, one or more selected from among: S102P, V147L, V147M, N154S, V155M, G166E, C206Y, G207E, S102P/F279L, F279L, R281Q, R284G, R284S, R284M, R284K, R284T, R197A, D205A, R310A, R293A, T294A, E296A, R197A/D205A, S272A/Q273A, R310A/E316A, E316A, E316N, E316Q, S272A, R293A/T294A/E296A, D231A, R232A, K236A, Q273A, S358A/E360A/S366A, D231A/R232A/K236A/R238A, S358A, E360A, S366A, R238A, R375A, N154S/R284G, and S324A/S326A; b) in MDA5, with reference to SEQ ID NO:310, one or more of: T331I, T331R, A489T, R822Q, G821S, A946T, R337G, D393V, G495R, R720Q, R779H, R779C, L372F, and A452T; c) in RIG-I, with reference to SEQ ID NO:311, one or both of E373A and C268F; and d) in IRF-3, with reference to SEQ ID NO:312, S396D, such as a variant STING that contains one or more amino replacement(s) selected, with reference to SEQ ID NOs: 305-309, from among: S102P, V147L, V147M, N154S, V155M, G166E, C206Y, G207E, S102P/F279L, F279L, R281Q, R284G, R284S, R284M, R284K, R284T, R197A, D205A, R310A, R293A, T294A, E296A, R197A/D205A, S272A/Q273A, R310A/E316A, E316A, E316N, E316Q, S272A, R293A/T294A/E296A, D231A, R232A, K236A, Q273A, S358A/E360A/S366A, D231A/R232A/K236A/R238A, S358A, E360A, S366A, R238A, R375A, N154S/R284G, and S324A/S326A, and conservative replacements thereof, and combinations thereof.

The immunostimulatory bacteria also can encode an immunostimulatory protein that confers or contributes to an anti-tumor immune response in the tumor microenvironment. These include, but are not limited to, a cytokine, a chemokine, or a co-stimulatory molecule. Exemplary of these is a protein selected from among one or more of: IL-2, IL-7, IL-12p70 (IL-12p40 + IL-12p35), IL-15, IL-36 gamma, IL-2 that has attenuated binding to IL-2Ra, IL-15/IL-15R alpha chain complex, IL- 18, IL-21, IL-23, IL-2 that is modified so that it does not bind to IL-2Ra, CXCL9, CXCL10, CXCL11, interferon-a, interferon-p, interferon-y, CCL3, CCL4, CCL5, proteins that are involved in or that effect or potentiate the recruitment and/or persistence of T cells, CD40, CD40 ligand (CD40L), CD28, 0X40, 0X40 ligand (OX40L), 4-1BB, 4- 1BB ligand (4-1BBL), members of the B7-CD28 family, CD47 antagonists, an anti- IL-6 antibody or an IL-6 binding decoy receptor, TGF-beta polypeptide antagonists, and members of the tumor necrosis factor receptor (TNFR) superfamily. The costimulatory molecule, selected from among CD40, CD40 ligand, CD28, 0X40, 0X40 ligand, 4-1BB, and 4-1BB ligand, can be truncated, such that the molecule lacks a cytoplasmic domain, or a portion thereof, for expression on an antigen-presenting cell (APC); and the truncated gene product is capable of constitutive immunostimulatory signaling to a T-cell through co-stimulatory receptor engagement, and is unable to counter-regulatory signal to the antigen-presenting cell (APC), due to the deleted, or partially deleted, or truncated cytoplasmic domain, which eliminates the immunosuppressive reverse signaling. Other such proteins are TGF-beta polypeptide antagonists, such as an anti-TGF-beta antibody or a fragment thereof, an anti-TGF- beta receptor antibody or a fragment thereof, a soluble TGF-beta antagonist polypeptide, or a TGF-beta binding decoy receptor.

The plasmids can encode a therapeutic antibody or antigen-binding fragment thereof, such as, for example, a Fab, Fab’, F(ab’)2, single-chain Fv (scFv), Fv, dsFv, nanobody, diabody fragment, or a single-chain antibody. Examples include, but are not limited to, an antagonist of PD-1, PD-L1, CTLA-4, VEGF, VEGFR2, or IL-6.

The plasmids can encode complementary products whose expression results in enhanced anti-tumor or other activity. For example, the combination of a modified, such as a constitutively active and/or chimeric STING protein described herein, with a cytokine, such as IL-15/IL-15R alpha chain complex (IL-15Ra-IL-15sc), has synergistic activity.

The immunostimulatory bacteria provided herein can be used for treatment of benign nervous system tumors. The tumors include, for example, wherein the subject is a subject having or diagnosed as having a benign tumor or tumor-associated condition selected from among neurofibromatosis 1 (NF1); neurofibromatosis 2 (NF2); schwannomatosis; meningioma; schwannoma; vestibular schwannoma; sporadic schwannoma, neurofibroma; neurofibromatosis (NF); and combinations thereof. Provided are methods of treatment and uses of the immunostimulatory bacteria provided herein for treating a subject having or at risk of having a benign nervous system tumor by using for treatment or administering to the subject a therapeutically effective amount of a composition comprising the immunostimulatory bacteria as described herein, such as those comprising the phenotype YS1646Aa«// AFLG/ SpagPi SansB! ScsgD or YS1646Aa«// AFLG/ SpagPI NcsgD . The bacteria optionally can be in combination with an immune checkpoint inhibitor, such as anti-PD-1 antibody or antagonist, or other checkpoint, and/or angiogenesis inhibitor. The nervous system tumors include schwannomas using attenuated Salmonella typhimurium and optionally one or more checkpoint inhibitors. Uses and methods using VNP20009 for treating such tumors are known (see, e.g., U.S. Publication No. 2022/0125906 and Ahmed etal. (2022) Proc. Natl. Acad. Sci. U.S.A. //9:e2202719119, which describes treatment with VNP20009). The properties of the immunostimulatory bacteria provided herein are superior to the VNP20009, as described throughout the disclosure herein, and thus, provide improved treatment for such conditions.

The bacteria provided herein, such as those comprising the phenotype YS1646Aa«// AFLG/ NpagP! SanslP ScsgD or YS 1646 Sasdl AFLG/ SpagP/ sgD, exhibit a broadly reduced systemic inflammatory signature, such as reduction in IL-2, TNF-alpha, IFN-gamma, IL-2, and IL-10, upon administration, as compared, for example, to the VNP20009 strain. The bacteria demonstrated safety up to the highest dose (3e ⁹ ) tested in a primate study, including low pro-inflammatory cytokines, and no anti-bacterial antibodies. These bacteria enrich in tumors and immunoprivileged tissue, are taken up by phagocytic cells, such as macrophages, and do not infect epithelial or endothelial cells. The bacteria can deliver complementary payload combinations, and are internalized by macrophages. If DNA is delivered it is expressed by proliferating macrophages. The bacteria exhibit significant T-cell infiltration in T-cell excluded tumors. Treated tumors show increases in activated CD8+ T-cells, decreases in exhausted T-cells, and Treg cells. The data in the working examples show cures in rodent models, including metastatic disease and protection from tumor re-challenge. Modified STING proteins and encoding nucleic acids

The delivery vehicles, including the immunostimulatory bacteria, can deliver nucleic acid encoding or protein that is part of a cytosolic DNA/RNA sensor pathway that leads to expression of type I interferon (IFN). These include STING, MDA5, IRF-3, IRF-7, IRF-5, IRF8, and RIG-I, and variants thereof, that have increased or constitutive activity in inducing type I interferon (IFN) upon infection of a cell, such as a macrophage. Also contemplated are delivery of an agonist of one or more of STING, MDA5, IRF-3, IRF-5, IRF-7, IRF-8, and/or RIG-I. In accord with methods and uses herein, the delivery vehicles can deliver DNA for transcription and translation in the eukaryotic host cell, and RNA and proteins as produced, for example, in the bacteria as described herein. Effecting the change in phenotype to the hybrid M1/M2 phenotype, can be accomplished by delivering bacteria, proteins, and RNA into the macrophage. For expression of proteins the macrophage are those that are proliferating.

Exemplary of such products included for delivery are modified STING proteins that have increased or constitutive activity whereby type I IFN expression is increased or constitutive. In general, the STING proteins comprise mutations whereby, when introduced into a eukaryotic cell, such as a human, type I IFN expression is constitutive. The STING proteins also can have lower NF-KB signaling activity than human STING. The STING proteins are provided in delivery vehicles, such as the bacteria provided herein, and other delivery vehicles, such as oncolytic vectors and nanoparticles, that encode the modified STING proteins for expression in subject to whom the delivery vehicle is administered.

Provided are the modified STING proteins and encoding nucleic acids, including plasmids and constructs, for expression thereof. Reference to the modified STING proteins includes reference to the encoding nucleic acids, plasmids, and constructs. Provided are modified Stimulator of Interferon Genes (STING) proteins from a non-human species, where the non-human STING is one that has lower NF-KB signaling activity compared to human STING, and, optionally, higher type I interferon (IFN) pathway signaling activity compared to human STING, where: the non-human STING protein is modified to include a mutation or mutations so that it has increased activity or acts constitutively in the absence of cytosolic nucleic acids; the mutations are insertions, deletions, and/or replacements of amino acids; and the STING protein optionally has a deletion or disruption of the TRAF6 binding site. Also provided are modified Stimulator of Interferon Genes (STING) proteins from a non-human species, or chimeric human STING proteins and modified forms thereof, comprising one or more mutation(s) associated with gain-of-function (GOF) that result in the constitutive activation of the encoded STING protein and/or enhanced sensitivity, or increased affinity or binding to endogenous ligands, whereby the STING protein is modified by one or more of an insertion, deletion, and replacement of an amino acid or amino acids; the STING protein has IFN-beta signaling activity, and attenuated nuclear factor kappa-light-chain-enhancer of activated B cell (NF-KB) signaling activity, compared to human STING; and the mutation or mutations result in increased STING activity or constitutive activity in inducing IFN-beta production. Human STING protein comprises the sequence set forth in any of SEQ ID N0s:305-309, or is a human allelic variant thereof with at least 98% sequence identity to the sequence of amino acids set forth in any of SEQ ID N0s:305-309.

The modified STING proteins include modified STING proteins, where: the STING protein is a chimera comprising replacement of a C-terminal tail (CTT) region in a STING protein from a first species, with the CTT of a STING protein from a second species; the STING protein of the second species has lower NF-KB signaling activity than the NF-KB signaling activity of human STING; and the TRAF6 binding site in the CTT optionally is deleted. Provided are the protein with the TRAF6 binding site, and without the TRAF6 binding site. Mutations that correspond to those that result in constitutive type I IFN expression include mutation or mutations is/are any that correspond to those associated with the auto-inflammatory disease STING- associated vasculopathy (SAVI) inhuman.

Provided are modified Stimulator of Interferon Genes (STING) proteins that are chimeras, comprising replacement of the CTT (C-terminal tail) region in a STING protein from a first species, with the CTT of a STING protein from a second species, where: the STING protein of the second species has lower NF-KB signaling activity than the NF-KB signaling activity of human STING; and the TRAF6 binding site in the CTT optionally is deleted. The chimeras can be a human STING protein with the replaced CTT, and optional TRAF6 binding site. Exemplary human STING protein comprises the sequence set forth in any of SEQ ID N0s:305-309, or is a human allelic variant thereof with at least 98% sequence identity to the sequence of amino acids set forth in any of SEQ ID N0s:305-309. For purposes of comparison when referencing human STING protein NF-KB signaling activity, the human STING protein has the sequence set forth in any of SEQ ID NOs. 305-309, if necessary to specify a particular allele, reference is to the protein of SEQ ID NO: 305. Exemplary chimeric STING proteins include those, where the first species is human, and the second species is selected from among Tasmanian devil, marmoset, cattle, cat, ostrich, boar, bat, manatee, crested ibis, coelacanth, and ghost shark. The chimeric STING proteins can include one or more of the mutations that render activity for inducing type I IFN expression constitutive.

In selecting a non-human STING protein for use in a chimera or for modification, the type I IFN signaling activity is at least or at least about 30%, 50%, 70%, 80% or more that of a wild type human STING protein, and generally is close to or higher than the human STING. The NF-KB signaling activity is less than 30%, less than 20%, less than 15%, less than 10%, or less than 5% that of wild type human STING NF-KB signaling activity. Exemplary of non-human species or second species is selected from among Tasmanian devil, marmoset, cattle, cat, ostrich, boar, bat, manatee, crested ibis, coelacanth, and ghost shark.

The modifications of STING are referenced by alignment with human STING of SEQ ID N0s:305-309, such as SEQ ID NO:305. Mutations that render a STING constitutive include, for example, a mutation or mutations that correspond, by reference to and alignment with human STING, to a mutation that occurs in an interferonopathy, wherein the sequence of human STING with which alignment is effected is set forth in any of SEQ ID N0s:305-309. Exemplary of such mutations is N154S, R284G, and N154S/R284G, and the others listed herein or known in the art. Exemplary of modified STING proteins are those that comprise replacement of the C- terminal tail (CTT) with the CTT from a STING protein that has reduced NF-KB signaling activity compared to the NF-KB signaling activity of human STING, such as where the replacing CTT is from a Tasmanian devil, marmoset, cattle, cat, ostrich, boar, bat, manatee, crested ibis, coelacanth, or ghost shark STING protein. Exemplary of replacing CTTs are any selected from among the following species Tasmanian devil, marmoset, cattle, cat, ostrich, boar, bat, manatee, crested ibis, coelacanth, or ghost shark STING protein, and it replaces the human STING CTT. Exemplary CTT sequences include those selected from among the following species and sequences: Tasmanian devil RQEEFAIGPKRAMTVTTSSTLSQEPQLLISGMEQPLSLRTDGF SEQ ID NO:371, Marmoset EEEEVTVGSLKTSEVPSTSTMSQEPELLISGMEKPLPLRSDLF SEQ ID NO:372, Cow EREVTMGSTETSVMPGSSVLSQEPELLISGLEKPLPLRSDVF SEQ ID NO:373, Cat EREVTVGSVGTSMVRNPSVLSQEPNLLISGMEQPLPLRTDVF SEQ ID NO:374, Ostrich RQEEYTVCDGTLCSTDLSLQISESDLPQPLRSDCL SEQ ID NO:375, Boar EREVTMGSAETSVVPTSSTLSQEPELLISGMEQPLPLRSDIF SEQ ID NO:376, Bat EKEEVTVGTVGTYEAPGSSTLHQEPELLISGMDQPLPLRTDIF SEQ ID NO:377, Manatee EREEVTVGSVGTSWPSPSSPSTSSLSQEPKLLISGMEQPLPLRTDVF SEQ ID NO:378, Crested ibis CHEEYTVYEGNQPHNPSTTLHSTELNLQISESDLPQPLRSDCF SEQ ID NO:379, Coelacantli

(vanant 1) QKEEYFMSEQTQPNSSSTSCLSTEPQLMISDTDAPHTLKRQVC SEQ ID NO:380, Coelacantli

(variant 2) QKEEYFMSEQTQPNSSSTSCLSTEPQLMISDTDAPHTLKSGF SEQ ID NO:381, and Ghost shark LTEYPVAEPSNANETDCMSSEPHLMISDDPKPLRSYCP SEQ ID NO:383, and allelic variants of each of these sequences, or variants having at least 98% sequence identity thereto.

The human CTT that can be replaced, comprises, for example, the sequence EKEEVTVGSLKTSAVPSTSTMSQEPELLISGMEKPLPLRTDFS (SEQ ID NO:370), or is an allelic or other variant having at least 98% sequence identity thereto. An exemplary chimeric STING is on that is a chimera in which the human STING CTT is replaced with a CTT from the Tasmanian devil STING. The chimera optionally includes a mutation or mutations rendering the type I IFN expression constitutive, where the STING protein is active in the absence of inducing ligands and/or inducing cytosolic nucleic acids. The replacing CTT from Tasmanian devil STING comprises the sequence: RQEEFAIGPKRAMTVTTSSTLSQEPQLLISGMEQPLSLRTDGF (SEQ ID NO:371), or is an allelic or other variant having at least 98% sequence identity thereto. The modified STING proteins optionally comprise deletion or disruption of the TRAF6 binding site, such as the TRAF6 binding site that comprises the amino acid residues DFS at the C-terminus as in human STING.

Modifications that render activity constitutive include, for example, one or more amino acid replacements that correspond(s) to one or more of S102P, V147L, V147M, N154S, V155M, G166E, C206Y, G207E, S102P/F279L, F279L, R281Q, R284G, R284S, R284M, R284K, R284T, R197A, D205A, R310A, R293A, T294A, E296A, R197A/D205A, S272A/Q273 A, R310A/E316A, E316A, E316N, E316Q, S272A, R293A/T294A/E296A, D231A, R232A, K236A, Q273A, S358A/E360A/S366A, D231A/R232A/K236A/R238A, S358A, E360A, S366A, R238A, R375A, and S324A/S326A, with reference, for alignment, to the sequence of human STING, as set forth in any of SEQ ID N0s:305-309. Exemplary of these are replacements corresponding to C206Y or R284G or N154S and combinations thereof, with reference to the sequence of human STING, as set forth in any of SEQ ID N0s:305-309.

Immunostimulatory bacteria provided herein can be used as a vaccine (and as a cancer therapeutic) by encoding an antigen against which an immune response, or immunization, or immuno-protection is desired. The immunostimulatory bacteria herein can be used to deliver RNA, such as mRNA or other forms, for use as a vaccine or for delivery of a therapeutic. As described herein, the bacteria contain a plasmid that encodes a product of interest, such as a therapeutic product, such as an antigen from a pathogen, under control of a bacterial or other prokaryotic promoter recognized by the bacterium. The encoding nucleic acid cassette includes a regulatory sequence or other sequence that blocks or inhibits or prevents translation by bacterial ribosomes, but that permits, provides for, or enhances translation by eukaryotic ribosomes, such as those present in human cells. The bacteria are modified so that they cannot grow or replicate in eukaryotes, such as by rendering the bacteria asd which require DAP for growth in vitro, or thyA', which requires thymidine monophosphate precursors, for growth, but can be cultured in vitro so that they produce the encoded RNA. A skilled person can inactivate the gene or product by modifying the endogenous gene, such as by deletions, insertions, replacements, transpositions, or any such modification so that active enzyme is not produced. See, SEQ ID NO:464 for an exemplary thyA gene from Salmonella, and SEQ ID NO:465 for the encoded protein. RNA, encoding protein and/or antigen for immunization, is encoded in the plasmid, but the encoding nucleic acid includes translation signals/sequences so that bacteria cannot translate the RNA. The resulting bacteria deliver the encoded RNA into the host phagocytic cells, where it is translated by host cell ribosomes. Immunostimulatory bacteria that are thyA' have genome modifications, such insertions, deletions, replacements, or other changes, that result in inactive or eliminate production of thymidylate synthase, which catalyzes the reductive methylation of dUMP to dTMP, a DNA biosynthesis precursor (precursor to dTTP).

AthyA auxotrophies for other nutrients and essential products can be introduced in place of or in addition to the asd inactivation/deletion. Other deletions or inactivation of genes or gene products required for growth, such as genes that produce nutrients, can be used in place of or in addition to the asd, and include for example thyA (see, e.g., Loessner et al. (2006) FEBS Lett 265:81-88). Elimination of expression or production or other attenuating mutations of the bacterial genome for production of such products results in release of encoded macromolecules upon bacterial cell death in vivo after administration. Asd is an essential enzyme for bacterial cell wall synthesis; ThyA is an enzyme needed for DNA synthesis. Mutation of the respective genes renders the strain auxotrophic for diaminopimelic acid (DAP) or thymidine monophosphate precursors. Upon deprivation of the complementing substrates, such bacteria die by DAP-less or thymine-less death, resulting in release of bacterial proteins and plasmid. Inactivation or elimination of Asd, results in release of macromolecules; elimination or inactivation of ThyA (to produce LthyA bacteria) expression/activity does not result in release of macromolecules, including proteins and plasmids, upon thymidine starvation (Leossner et al. (2006) FEBS Lett 265:81- 88). Thus, AthyA are advantageous for in vivo delivery of plasmids to host cells, since the bacteria will not prematurely release their contents. Since the bacteria provided here infect or accumulate in myeloid cells, such as phagocytic cells, such as macrophages, dendritic cells, monocytes and neutrophils, which consume bacteria, the intact EthyA bacteria, release the plasmid encoding the therapeutic product inside the targeted cells.

The bacteria are genome-modified so that they are attenuated, such as the bacteria herein, where the response by toll-like receptors (TLR) 2, 4, and 5 is reduced, compared to bacteria without such genome modifications, and optionally, encode rck (resistance to complement killing) to reduce inactivation by complement, and include modifications, as needed, so that they infect primarily or only phagocytic cells, such as tissue-resident macrophages. It is shown herein that genome modifications, such the combination of modifications that reduce responses by TLRs 2,4,5, are necessary for the production of type I IFN by human antigen-presenting cells.

Provided are bacteria that contain genome modifications, whereby the response by toll-like receptors (TLRs) 2, 4, and 5 is reduced, compared to the bacteria without the genome modifications. Such modifications include those that result in penta-acylated LPS and elimination of flagella, such as the pagP~!msbB~ bacteria that lack flagella, and also those that are deficient or do not produce or express asparaginase II, such as those that are EansB. The bacteria also can comprise further genomic modifications such as one or more modifications whereby they are

- 83 -

RECTIFIED SHEET (RULE 91 ) ISA/EP auxotrophic for a required nutrient or for a factor, so that they are unable to replicate in a eukaryotic host, but can replicate in vitro when supplied with the nutrient or factor, such auxotrophic for thymidine (tsihyA), such as by genome modifications that render them unable to produce or express thymidylate synthase (AthyA), or Asd.

The bacteria that are provided herein that combine some or all of these traits are used to express therapeutic products, including anti-cancer products, and antigens, depending upon their intended use. For administration to subjects with cancer, the bacteria, which accumulate in tumor-resident myeloid cells, encode anti-cancer therapeutics, such as products that result in stimulation of an immune response, and/or that result in inhibiting immunosuppression, or that encode a product that treats the tumor, and those that encode a combination of products that can act synergistically to treat cancer. The bacteria provided herein that accumulate in or infect phagocytic cells, also can be used for subjects that do not have cancer, such as, as vaccines by delivering or encoding an antigen, or delivering RNA. The various embodiments and combinations of properties and products, and uses, are described throughout the disclosure herein.

In some embodiments, the bacteria comprise a plasmid containing nucleic acid encoding a product, or comprise RNA encoding the product, where the product encoded by the nucleic acid or RNA is an antigenic sequence or sequences from a pathogenic virus, bacterium, parasite, or is a tumor antigen, whereby, upon expression of the encoded antigen in the host, the host develops an immune-protective response or immunizing response against the pathogenic virus, bacterium, parasite, or tumor antigen, or the encoded product is a therapeutic product; expression of the antigenic sequence(s) is/are under control of a prokaryotic promoter so that RNA encoding the antigen(s) is/are produced in the bacteria; nucleic acid encoding the antigen comprises regulatory sequences that inhibit or prevent translation of encoded RNA by bacterial ribosomes, but that does not inhibit or prevent translation of the encoded RNA by eukaryotic host ribosomes, whereby translation is de-coupled from transcription in the bacteria; the resulting bacteria are selective for infecting phagocytic cells when administered to a eukaryotic subject, and deliver the nucleic acid into the phagocytic cells, wherein the RNA is translated.

The bacteria, which are cultured in vitro to produce the RNA, upon administration, infect the phagocytes and deliver their contents, but they are not viable and/or do not replicate, thereby providing the RNA, such as mRNA, to the host cells, which translate the RNA to produce the encoded product, such as an immunogenic protein or antigen. The RNA generally is mRNA, and also can be other forms of RNA, such as RNAi, or eRNA (circular RNA), and other therapeutic forms. The immunostimulatory bacteria that are used for this purpose can include the plasmids, which encode the RNA, in high or higher (generally 150 or greater) copy numbers, to increase the amount of RNA delivered. Various embodiments are described, claimed, and exemplified herein. The mRNA can encode pathogen proteins, pathogen antigens, tumor-antigens, therapeutic products for treatment of tumors or infections, and combinations thereof. The mRNA can be synthetic, such as those designed for immunization (see, e.g., US patent publication 20190351040, and others that describe mRNA for immunization or treatment). The resulting bacteria are vaccines for therapy or immunization. The payloads can include products that are adjuvants, that are immunostimulatory protein, that induce type I interferon (IFN) to activate T-cells in concert with the immunizing antigen/protein.

In some embodiments, the immunostimulatory bacteria provided herein contain a plasmid that encodes two or more therapeutic proteins selected from among: a) an immunostimulatory protein that confers or contributes to an anti-tumor immune response in the tumor microenvironment; b) one or more of a protein that is part of a cytosolic DNA/RNA sensor pathway that leads to expression of type I interferon (IFN), or a variant thereof that has increased activity to increase expression of type I IFN, or a variant thereof that results in constitutive expression of a type I IFN; and c) an anti-cancer antibody or antigen-binding portion thereof. For example, the immunostimulatory protein can be a co-stimulatory molecule that is one that lacks a cytoplasmic domain or a sufficient portion thereof, for expression on an antigen- presenting cell (APC), whereby the truncated co-stimulatory molecule is capable of constitutive immunostimulatory signaling to a T-cell through co-stimulatory receptor engagement, and is unable to counter-regulatory signal to the antigen presenting cell (APC). In some embodiments, the immunostimulatory bacteria encode at least two therapeutic products, selected from among a cytokine, a protein that constitutively induces a type I IFN, a co-stimulatory molecule, and an anti-cancer antibody or antigen-binding portion thereof, which can be under control of a single promoter. For example, expression of the nucleic acid encoding at least two or all of the products is under control of a single promoter, and the nucleic acid encoding each product is separated by nucleic acid encoding 2A polypeptides, whereby, upon translation, each product is separately expressed. The nucleic acid encoding each product can be operatively linked to nucleic acid encoding a sequence that directs secretion of the expressed product from a cell.

Provided are immunostimulatory bacteria that encode two or more therapeutic products, wherein at least one product is selected from a), and at least one is selected from b), and a) is IL-2, IL-7, IL-12p70 (IL-12p40 + IL-12p35), IL-15, IL-23, IL-36 gamma, IL-2 that has attenuated binding to IL-2Ra, IL-15/IL-15R alpha chain complex (also referred to herein as IL-15/IL-15Ra, IL-15 complex, or other variations), IL-18, IL-2 that is modified so that it does not bind to IL-2Ra, CXCL9, CXCL10, CXCL11, interferon-a, interferon- , CCL3, CCL4, CCL5, proteins that are involved in or that effect or potentiate the recruitment and/or persistence of T cells, CD40, CD40 ligand (CD40L), 0X40, 0X40 ligand (OX40L), 4-1BB, 4-1BB Ligand (4-1BBL), members of the B7-CD28 family, TGF-beta polypeptide antagonists, or members of the tumor necrosis factor receptor (TNFR) superfamily; and b) is STING, RIG-I, MDA-5, IRF-3, IRF-5, IRF-7, TRIM56, RIP1, Sec5, TRAF3, TRAF2, TRAF6, STAT1, LGP2, DDX3, DHX9, DDX1, DDX9, DDX21, DHX15, DHX33, DHX36, DDX60, and SNRNP200. They also can encode one or more of a TGF-beta inhibitory antibody, a TGF-beta binding decoy receptor, an anti-IL6 antibody, or an IL-6 binding decoy receptor.

Exemplary of combinations of encoded therapeutic products are any of the following combinations of therapeutic products: IL-2 and IL-12p70; IL-2 and IL-21; IL-2, IL-12p70, and a STING GOF variant; IL-2, IL-21, and a STING GOF variant; IL-2, IL-12p70, a STING GOF variant, and 4-1 BBL (including 4-lBBLAcyt), where Acyt is a deleted cytoplasmic domain; IL-2, IL-21, a STING GOF variant, and 4- 1BBL (including 4-lBBLAcyt); IL-15/IL-15Ra, and a STING GOF variant; IL-15/IL- 15Ra, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); IL-15/IL-15Ra and IL-12p70; IL-15/IL-15Ra and IL-21; IL-15/IL-15Ra, IL-12p70, and a STING GOF variant; IL-15/IL-15Ra, IL-21, and a STING GOF variant; IL-15/IL-15Ra, IL- 12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); IL-15/IL-15Ra, IL-21, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); IL-12p70 and IL-21; IL-12p70, IL-21, and a STING GOF variant; IL-12p70, IL-21, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); IL-12p70 and a STING GOF variant; IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); IL-12p70 and IL-18; IL-12p70, IL-18, and a STING GOF variant; IL-12p70, IL-18, a STING GOF vanant, and 4-1 BBL (including 4-lBBLAcyt); a TGF-P decoy receptor, IL-2, and IL-12p70; a TGF-P decoy receptor, IL-2, and IL-21; a TGF-P decoy receptor, IL- 2, IL-12p70, and a STING GOF variant; a TGF-P decoy receptor, IL-2, IL-21, and a STING GOF variant; a TGF-P decoy receptor, IL-2, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a TGF-P decoy receptor, IL-2, IL-21, a STING GOF variant, and 4-1 BBL (including 4-lBBLAcyt); a TGF-P decoy receptor, IL- 15/IL-15Ra, and a STING GOF variant; a TGF-P decoy receptor, IL-15/IL-15Ra, a STING GOF variant, and 4-1 BBL (including 4-lBBLAcyt); a TGF-P decoy receptor, IL-15/IL-15Ra, and IL-12p70; a TGF-P decoy receptor, IL- 15/IL- 15Ra, and IL-21; a TGF-P decoy receptor, IL-15/IL-15Ra, IL-12p70, and a STING GOF variant; a TGF- P decoy receptor, IL-15/IL-15Ra, IL-21, and a STING GOF variant; a TGF-P decoy receptor, IL- 15/IL- 15Ra, IL-12p70, a STING GOF variant, and 4-1BBL (including 4- IBBLAcyt); a TGF-P decoy receptor, IL-15/IL-15Ra, IL-21, a STING GOF variant and, 4-1BBL (including 4-lBBLAcyt); a TGF-p decoy receptor, IL-12p70, and IL-21; a TGF-P decoy receptor, IL-12p70, IL-21, and a STING GOF variant; a TGF-P decoy receptor, IL-12p70, IL-21, a STING GOF variant, and 4-1BBL (including 4- IBBLAcyt); a TGF-P decoy receptor and IL-12p70; a TGF-p decoy receptor, IL- 12p70, and a STING GOF variant; a TGF-P decoy receptor, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a TGF-P decoy receptor, IL-12p70, and IL-18; a TGF-P decoy receptor, IL-12p70, IL-18, and a STING GOF variant; a TGF-P decoy receptor, IL-12p70, IL-18, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a TGF-P decoy receptor and a STING GOF variant; an anti- CTLA-4 antibody, IL-2, and IL-12p70; an anti-CTLA-4 antibody, IL-2, and IL-21; an anti-CTLA-4 antibody, IL-2, IL-12p70, and a STING GOF variant; an anti-CTLA-4 antibody, IL-2, IL-21, and a STING GOF variant; an anti-CTLA-4 antibody, IL-2, IL- 12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); an anti-CTLA-4 antibody, IL-2, IL-21, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); an anti-CTLA-4 antibody, IL-15/IL-15Ra, and a STING GOF variant; an anti-CTLA- 4 antibody, IL-15/IL-15Ra, a STING GOF variant, and 4-1BBL (including 4- IBBLAcyt); an anti-CTLA-4 antibody, IL-15/IL-15Ra, and IL-12p70; an anti-CTLA- 4 antibody, IL-15/IL-15Ra, and IL-21; an anti-CTLA-4 antibody, IL-15/IL-15Ra, IL- 12p70, and a STING GOF variant; an anti-CTLA-4 antibody, IL-15/IL- 15Ra, IL-21, and a STING GOF variant; an anti-CTLA-4 antibody, IL-15/IL-15Ra, IL-12p70, a STING GOF variant, and 4-1 BBL (including 4-lBBLAcyt); an anti-CTLA-4 antibody, IL-15/IL-15Ra, IL-21, a STING GOF vanant, and 4-1BBL (including 4- IBBLAcyt); an anti-CTLA-4 antibody, IL-12p70, and IL-21; an anti-CTLA-4 antibody, IL-12p70, IL-21, and a STING GOF variant, an anti-CTLA-4 antibody, IL- 12p70, IL-21, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); an anti- CTLA-4 antibody and IL-12p70; an anti-CTLA-4 antibody, IL-12p70, and a STING GOF variant; an anti-CTLA-4 antibody, IL-12p70, a STING GOF variant, and 4- 1BBL (including 4-lBBLAcyt); an anti-CTLA-4 antibody, IL-12p70, and IL-18; an anti-CTLA-4 antibody, IL-12p70, IL-18, and a STING GOF variant; an anti-CTLA-4 antibody, IL-12p70, IL-18, a STING GOF variant, and 4-1BBL (including 4- IBBLAcyt); an anti-CTLA-4 antibody and a STING GOF variant; a CD40 agonist, IL-2, and IL-12p70; a CD40 agonist, IL-2, and IL-21; a CD40 agonist, IL-2, IL- 12p70, and a STING GOF variant; a CD40 agonist, IL-2, IL-21, and a STING GOF variant; a CD40 agonist, IL-2, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a CD40 agonist, IL-2, IL-21, a STING GOF variant, and 4- 1BBL (including 4-lBBLAcyt); a CD40 agonist, IL- 15/IL- 15Ra, and a STING GOF variant; a CD40 agonist, IL-15/IL-15Ra, a STING GOF variant, and 4-1 BBL (including 4-lBBLAcyt); a CD40 agonist, IL-15/IL-15Ra, and IL-12p70; a CD40 agonist, IL-15/IL-15Ra, and IL-21; a CD40 agonist, IL-15/IL-15Ra, IL-12p70, and a STING GOF variant; a CD40 agonist, IL-15/IL-15Ra, IL-21, and a STING GOF variant; a CD40 agonist, IL-15/IL-15Ra, IL-12p70, a STING GOF variant, and 4- 1BBL (including 4-lBBLAcyt); a CD40 agonist, IL- 15/IL- 15Ra, IL-21, a STING GOF variant, and 4-1 BBL (including 4-lBBLAcyt); a CD40 agonist, IL-12p70, and IL-21; a CD40 agonist, IL-12p70, IL-21, and a STING GOF variant; a CD40 agonist, IL-12p70, IL-21, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a CD40 agonist and IL-12p70; a CD40 agonist, IL-12p70, and a STING GOF variant; a CD40 agonist, IL-12p70, a STING GOF variant, and 4-1BBL (including 4- IBBLAcyt); a CD40 agonist, IL-12p70, and IL-18; a CD40 agonist, IL-12p70, IL-18, and a STING GOF variant; a CD40 agonist, IL-12p70, IL-18, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); and a CD40 agonist and a STING GOF variant.

Other combinations of products include, for example, IL-15 and a STING gain-of-function variant, including STING chimeras with a gain-of-function mutation or mutations, as provided herein, or IL-15Ra-IL-15sc and a STING gain-of-function variant, including STING chimeras with a gain-of-function mutation or mutations. Other products or combinations thereof include a bi-specific T-cell engager (BiTe®); a BiTe® and a STING protein, such as a modified GOF STING protein or a STING chimera, as described herein; a BiTe® and IL-15; a BiTe® and IL-15Ra-IL-15sc; a BiTe®, IL-15 and a STING protein, such as a modified GOF STING protein or chimeric STING protein; and a BiTe®, IL-15Ra-IL-15sc, and a STING protein, such as a modified GOF STING protein or chimeric STING protein, where the BiTe® targets include, for example, DLL3, EGFR, Her2, CEA, Mesothelin, PSMA, EpCAM, CD74, Folate Receptor, Nectin4, EphA2, CA-IX, B7H3, Siglec-15, Mucl, Lewis Y antigen, and other such tumor antigens/ tumor targets.

Also provided are therapeutic compositions containing a tumor antigen(s) and a STING gain-of-function variant or STING chimera; a therapeutic composition of a tumor antigen(s) and IL- 15; a therapeutic composition of a tumor antigen(s) and IL- 15Ra-IL-15sc; a therapeutic composition of a tumor antigen(s), IL-15, and a STING gain-of-function variant or STING chimera; and a therapeutic composition of a tumor antigen(s), IL-15Ra-IL-15sc, and a STING gain-of-function variant or STING chimera. These products can be encoded in the immunostimulatory bacteria. The tumor antigens can be any listed or described herein (for example, in Example 35), or known in the art.

Combinations of products also include combinations of antigens and immune stimulating proteins. The antigens can be tumor antigens, or they can be immunizing antigens, such as pathogenic antigens, where the pathogens include, for example, bacteria, protozoans, viruses, and prions, and other prion-like particles that cause diseases and disorders. The antigens include any described or listed herein, or known in the art. The combinations include, for example, combinations of one or more antigens and IFNa2; one or more antigens and IFN-P; one or more antigens, IFNa2, and IFN-P; one or more antigens and an IRF3 GOF variant with the mutation S396D; and one or more antigens, IFNa2, and an IRF3 GOF variant with the mutation S396D.

Other products and combinations of products that are encoded in the immunostimulatory bacteria provided herein, include, but are not limited to, the combination of IFNa2 and an IRF3 GOF variant with the mutation S396D; IFNa2 and IFN-P; FLT-3L (FMS-like tyrosine kinase 3 ligand; see, e.g., SEQ ID NO:436); sialidase (see, e.g., SEQ ID NO:435); the IL-12 p35 subunit of IL-12p70 only; Azurin; a membrane anchored/tethered cytokine or molecule, such as, for example, IL-2, IL-12, IL-12p35, IL-21, IL-15, IL-15Ra-IL-15sc, or FLT-3L; or a TLR8 agonist, such as, for example, where the TL38 agonist is polyU or polyU/G, a microRNA, or is miR-21.

Also provided are modified non-human Stimulator of Interferon Genes (STING) proteins, and STING protein chimeras, as well as delivery vehicles, including any described herein, pharmaceutical compositions, cells encoding or containing these STING proteins, and uses thereof, and methods of treatment of cancers. In particular, the immunostimulatory bacteria provided herein encode the modified non-human STING proteins, non-human STING proteins, and STING chimeras, as described herein. These STING proteins that are encoded by the immunostimulatory bacteria are provided herein and described throughout.

Provided herein are modified non-human STING proteins, where the non- human STING protein is one that has lower F-KB activation than the human STING protein, and, optionally, higher type I interferon activation activity compared to the wild-type (WT) human STING protein. These non-human STING proteins are modified to include a mutation or mutations so that they have increased activity, or act constitutively, in the absence of cytosolic nucleic acid signaling. The mutations are typically amino acid mutations that occur in interferonopathies in humans, such as those described above for human STING. The corresponding mutations are introduced into the non-human species STING proteins, where corresponding amino acid residues are identified by alignment. Also, in some embodiments, the TRAF6 binding site in the C-terminal tail (CTT) of the STING protein is deleted, reducing NF-KB signaling activity.

Provided are modified STING proteins, particularly human STING proteins, that are chimeras, in which the CTT (C-terminal tail) region in the STING protein from one species, such as human, is replaced with the CTT from a STING protein of another species that has lower NF-KB signaling activity and/or higher type I IFN signaling activity than human STING. Also, the TRAF6 binding site is optionally deleted in these chimeras.

The modified STING proteins also include the mutations as set forth throughout the disclosure herein.

Also provided are delivery vehicles, such as immunostimulatory bacteria, and any provided herein or known to those of skill in the art, including, for example, exosomes, nanoparticles, minicells, cells, liposomes, lysosomes, oncolytic viruses, and other viral vectors, that encode the modified STING proteins of any of 1-3. Also provided are delivery vehicles, such as immunostimulatory bacteria, any provided herein or known to those of skill in the art, including, for example, exosomes, nanoparticles, minicells, cells, liposomes, lysosomes, oncolytic viruses, and other viral vectors, that encode unmodified STING from a non-human species whose STING protein has reduced F-KB signaling activity compared to that of human STING, and optionally, increased type I interferon stimulating/signaling activity compared to that of human STING.

Also provided are cells (non-zygotes, if human), such as cells used for cell therapy, such as T-cells and stem cells, and cells used to produce the STING proteins as described herein. Also provided are pharmaceutical compositions that contain the STING proteins, or the delivery vehicles, or the cells, or combinations thereof.

Uses and methods of treatment of cancer and vaccination against pathogens or cancer by administering any the immunostimulatory bacteria as described herein are provided.

Assays and methods to assess NF-KB activity (signaling activity), and type I interferon stimulating activity or interferon-P stimulating activity of STING are described herein, and also, are known to those of skill in the art. Methods include those described, for example, in de Oliveira Mann et al. (2019) Cell Reports 27'.1165- 1175, which describes, inter alia the interferon-p and NF-KB signaling activities of STING proteins from various species, including human, thereby identifying STING proteins from various species that have lower NF-KB activity than human STING, and those that also have comparable or higher interferon-P activity than human STING, de Oliveira Mann et al. (2019) provides species alignments and identifies domains of STING in each species, including the CTT domain (see, also, the Supplemental Information for de Oliveira Mann etal. (2019)).

The non-human STING proteins can be, but are not limited to, STING proteins from the following species: Tasmanian devil (Sarcophilus harrisii,' SEQ ID NO:349), marmoset (Callithrix jacchus,' SEQ ID NO:359), cattle (Bos taurus, SEQ ID NO:360), cat (Felis catiis, SEQ ID NO:356), ostrich (Struthio camelus australis,' SEQ ID NO:361), crested ibis (Nipponia nippon, SEQ ID NO:362), coelacanth (Latimeria chalumnae,' SEQ ID NOs:363-364), boar (Sus scrofa, SEQ ID NO:365), bat (Rousettus aegyptiacus,' SEQ ID NO:366), manatee (Trichechus manatus latirostris,' SEQ ID NO:367), ghost shark (Callorhinchus milii, SEQ ID NO:368), and mouse (Mus musculus. SEQ ID NO:369). These vertebrate STING proteins readily activate immune signaling in human cells, indicating that the molecular mechanism of STING signaling is shared in vertebrates (see, de Oliveira Mann et al. (2019) Cell Reports 27:1165-1175).

It is shown herein that the immunostimulatory bacteria provided herein, by virtue of the ability to infect myeloid cells, such as tumor-resident and tissue-resident macrophages, and the ability to retain viability for at least a limited time, and/or to deliver plasmids that encode therapeutic products that result in expression of type I IFN and/or other immune-stimulating products, such as gain-of-function (GOF) variants that do not require cytosolic nucleic acids, nucleotides, dinucleotides, or cyclic dinucleotides (CDNs) to result in expression of type I IFN, can convert macrophages that have the M2 phenotype into Ml or Ml -like macrophages, with immunosuppressive properties reduced or eliminated, and immune-stimulating, antitumor or anti-viral properties enhanced or added. Provided are immunostimulatory bacteria that contain a plasmid encoding a therapeutic product, where infection of a macrophage, including human macrophages, by the bacterium, converts an M2 macrophage to an Ml phenotype or Ml -like phenotype macrophage. Provided are immunostimulatory bacteria that contain a plasmid encoding a therapeutic product whose expression in a macrophage results in the conversion of, or converts, M2 macrophages, such as human M2 macrophages, to an Ml or Ml-like phenotype. The immunostimulatory bacteria with such properties include any of the bacteria provided herein that contain genome modifications that result in infection of tumor-resident (in subjects with cancer), and tissue-resident myeloid cells. These genome modifications include those that result in a lack of flagella, wherein the wild-type bacterium has flagella, and others, such as those that result in bacteria that are pagP' ImsbB' . Other modifications include those that result in the elimination of asparaginase activity, such as modifications that result in bacteria that are ansB', in the bacteria that infect myeloid cells, which thereby enhances T-cell activities, and other modifications that alter the lipopolysaccharide (LPS). These immunostimulatory bacteria provided herein convert immunosuppressive phagocytic macrophages to immunostimulatory, phagocytic macrophages that are capable of in situ antigen cross-presentation to CD8+ T-cells, and of migration to lymph nodes to prime CD4+ and CD8+ T-cells.

Included are immunostimulatory bacteria that encode therapeutic products in macrophages, that facilitate or result in the conversion of, or that convert, M2 macrophages to an Ml or Ml-like phenotype, which has a profile of some or all characteristics of Ml macrophage. Exemplary of the therapeutic products are those that are part of a cytosolic DNA/RNA sensor pathway that leads to expression of type I interferon (IFN), particularly constitutive expression. This includes the gain-of- function (GOF) variants of therapeutic products that are part of the cytosolic DNA/RNA sensor pathway, and that do not require cytosolic nucleic acids, nucleotides, dinucleotides, or cyclic dinucleotides (CDNs) to result in expression of type I IFN, such as the variant and non-human STING proteins, STING chimeras, and STING chimeras with gain-of-function mutations, as described and provided herein. The bacteria include any that can be modified as described herein, including the species listed herein, such as Salmonella species and strains.

Also provided are immunostimulatory bacteria that contain nucleic acid operatively linked to a prokaryotic promoter, where: the nucleic acid comprises RNA that lacks sequences necessary for translation by a prokaryotic cell, whereby the RNA is produced in the bacterium, but is not translated into protein. For example, the RNA lacks a Shine-Dalgamo sequence, and comprises an Internal Ribosome Entry Site (IRES) and/or a translational read-through 2A peptide. The IRES sequence prevents translation by prokaryotic ribosomes, but provides for translation by eukaryotic ribosomes. The bacteria include immunostimulatory bacteria in which the 2A peptide is one or more of T2A, P2A, E2A, or F2A, to produce discrete products from polycistronic constructs.

Also provided are immunostimulatory bacteria as described herein that can be a delivery vehicle for delivering RNA to eukaryotic cells, such as myeloid cells. These bacteria include nucleic acid operatively linked to a prokaryotic promoter, where: the nucleic acid and prokaryotic promoter generally are encoded on a plasmid, but in some embodiments, are encoded in the bacterial genome; the nucleic acid comprises RNA that lacks sequences necessary for translation by the bacterial ribosomes, whereby the RNA is produced in the bacterium, and where: the RNA lacks a Shine-Dalgarno sequence, and comprises an Internal Ribosome Entry Sequence (IRES), or a translational read-through 2A peptide. The prokaryotic promoter, when operatively linked to nucleic acid encoding a therapeutic protein (or non-bacterial protein), can be a bacterial promoter or a phage promoter, such as a bacteriophage promoter. The RNA polymerase that recognizes the phage promoter can be encoded in the bacterial genome, or on a plasmid for expression in the bacteria. Exemplary prokaryotic promoters include any known to those of skill in the art, including, but not limited to, those that comprise all or a sufficient portion of (sufficient to initiate transcription of an operatively linked nucleic acid) the promoters whose sequences are set forth in any of SEQ ID NOs: 393-396, respectively: attatgtcttgacatgtagtgagtgggctggtataatgcagcaag (SEQ ID NO:393), ttatgcttgacgctgcgtaaggtttttgttataatacaccaag (SEQ ID NO:394), or attatgtcttgacatgtagtgagtgggctggtaaatgcagcaag (SEQ ID NO: 395), or gatcccggagttcatgcgtgatgcaatgaaagtgccgttctacttcggtgggacctcact gcttatcgttgttgtcgtgattatg gactttatggctcaagtgcaaactctgatgatgtccagtcagtatgagtctgcattgaag aaggcgaacctgaaaggctacg gccgttaattggtcgcctgagaagttacggagagtaaaaatgaaagttcgtgcttccgtc aagaaattatgccgtaactgcaa aatcgttaagcgtgatggtgtcatccgtgtgatttgcagtgccgagccgaagcataaaca gcgccaaggctgattttttcgca tatttttcttgcaaagttgggttgagctggctagattagccagccaatcttttgtatgtc tgtacgtttccatttgagtatcctgaaa acgggcttttcagcatggtacgtacatattaaatagtaggagtgcatagtggcccgtata gcaggcattaacattcctgatca gaaacacgccgtgatcgcgttaacttcgatctacggtgtcggcaagacccgttctaaagc catcctggctgcagcgggtat cgctgaaaatgttaagatcctctagatttaagaaggagatatacat (Salmonella rpsM promoter; SEQ ID NO:396).

These immunostimulatory bacteria comprise genomic modifications, as described herein, whereby the bacteria infect tissue-resident myeloid cells, and/or tumorresident myeloid cells, or, in subjects that do not have tumors, in phagocytic cells, such as macrophages. The bacteria infect the cells and deliver the RNA, which is translated in the eukaryotic host cell. Exemplary of such bacteria, are those that are modified to lack flagella, such as by deletion or disruption of the genes involved in producing flagella. The bacteria are species and strains that, without the genomic modifications, have flagella.

Also provided are immunostimulatory bacteria in which an encoded therapeutic product, such as a protein, is linked to a moiety that confers an improved pharmacological property, such as a pharmacokinetic or pharmacodynamic property, such as increased serum half-life. Hence, provided are immunostimulatory bacteria, where an encoded therapeutic product comprises an Fc domain, or a half-life extending moiety, such as human serum albumin, or a portion thereof. Half-life extension modalities or methods, include, for example, PEGylation, modification of glycosylation, sialylation, PASylation (modification with polymers of PAS amino acids that are about 100-200 residues in length), ELPylation (see, e.g., Floss et al. (2010) Trends Biotechnol ,28(1):37 -45), HAPylation (modification with a glycine homopolymer), fusion to human serum albumin, fusion to GLK, fusion to CTP, GLP fusion, fusion to the constant fragment (Fc) domain of a human immunoglobulin (IgG), fusion to transferrin, fusion to non-structured polypeptides, such as XTEN (also referred to as rPEG, which is a genetic fusion of non-exact repeat peptide sequences, containing A, E, G, P, S, and T; see, e.g., Schellenberger et al. (2009) Nat. Biotechnol. 27(12): 1186-1190), and other such modifications and fusions that increase the size, increase the hydrodynamic radius, alter the charge, or target to receptors for recycling rather than clearance, and combinations of such modifications and fusions.

Also provided are immunostimulatory bacteria, where the encoded therapeutic product comprises the B7 protein transmembrane domain, or where the therapeutic product is GPI-anchored by virtue of an endogenous or added GPI anchor. The encoded therapeutic product can comprise a fusion to collagen.

The immunostimulatory bacteria in any and all embodiments can be any suitable species. Where reference is made to particular genes and gene modifications, the genes and modifications are those that correspond to the genes and modifications referenced with respect to Salmonella, as an exemplary species. Species and strains include, for example, a strain of Rickettsia, Klebsiella, Bordetella, Neisseria, Aeromonas, Francisella, Corynebacterium, Citrobacter, Chlamydia, Haemophilus, Brucella, Mycobacterium, Mycoplasma, Legionella, Rhodococcus, Pseudomonas, Helicobacter, Vibrio, Bacillus, and Erysipelothrix. For example, Rickettsia rickettsiae, Rickettsia prowazekii, Rickettsia tsutsugamuchi, Rickettsia mooseri, Rickettsia sibirica, Bordetella bronchiseptica, Neisseria meningitidis, Neisseria gonorrhoeae, Aeromonas eucrenophila, Aeromonas salmonicida, Francisella tularensis, Corynebacterium pseudotuberculosis, Citrobacter freundii, Chlamydia pneumoniae, Haemophilus somnus, Brucella abortus, Mycobacterium intracellulare , Legionella pneumophila, Rhodococcus equi, Pseudomonas aeruginosa, Helicobacter mustelae, Vibrio cholerae, Bacillus subtilis, Erysipelothrix rhusiopathiae, Yersinia enter ocolitica, Rochalimaea quintana, and Agrobacterium tumerfacium.

Provided herein are genome modified bacteria, comprising genome modifications, whereby TLR2, TLR4, and TLR5 signaling is reduced compared to the bacteria without the genome modifications, wherein: the bacteria comprise further genomic modifications whereby they are auxotrophic for a required nutrient or factor so that they are unable to replicate in a eukaryotic host, but can replicate in vitro when supplied with the nutrient or factor; the bacteria comprise a plasmid containing nucleic acid, or comprise RNA encoding an antigenic sequence or sequences from a pathogenic virus, bacterium, or parasite, or encoding a tumor antigen, whereby, upon expression of the encoded antigen in the host, the host develops an immmunoprotective response against the pathogenic virus, bacterium, or parasite; expression of the antigenic sequence(s) is/are under control of a prokaryotic promoter so that RNA encoding the antigen(s) is produced in the bacteria; nucleic acid encoding the antigen comprises regulatory sequences that inhibit or prevent translation of encoded RNA by bacterial ribosomes, but that does not inhibit or prevent translation of the encoded RNA by eukaryotic host ribosomes, whereby translation is de-coupled from transcription in the bacteria; the resulting bacteria infect phagocytic cells when administered to a eukaryotic subject, and deliver the nucleic acid into the phagocytic cells, wherein the RNA is translated.

The nucleic acid encoding the antigenic sequence(s) can comprise an internal ribosomal entry site (IRES) sequence, whereby host cell translation is facilitated or enhanced and bacterial translation is inhibited or prevented. The IRES can be Vascular Endothelial Growth Factor and Type 1 Collagen Inducible Protein (VCIP; see, e.g., SEQ ID NO:434), and the nucleic acid encoding the antigen(s) can comprise a VCIP IRES or other IRES that inhibits bacterial translation. The IRES or the VCIP IRES can be included in the plasmid, at a position that is 3’ of the promoter and 5’ of the antigen(s) coding sequence.

The pathogen can be a bacterium or a virus, or the encoded antigen can be a tumor antigen. The immunostimulatory bacteria provided herein can be vaccines to prevent or treat a viral infection or a bacterial infection, including chronic viral infections, and acute infections. The infection can be from an infection by hepatitis viruses, herpesviruses, varicella zoster virus (VZV), Epstein-Barr virus, human immunodeficiency virus (HIV), human T-cell leukemia virus (HTLV), Respiratory Syncytial Virus (RSV), measles virus, and other viruses that chronically infect subjects. The infectious agent can be Severe Acute Respiratory Syndrome coronavirus (SARS-CoV), Middle East Respiratory Syndrome coronavirus (MERS-CoV), or Severe Acute Respiratory Syndrome coronavirus 2 (SARS-CoV-2, which causes COVID-19).

The pathogen can be a species of Escherichia, Staphylococcus, Pseudomonas, or Porphyromonas, or the pathogen can be P. gingivalis, SARS-CoV, or E. coli. The plasmid in these immunostimulatory bacteria can further encode an immunostimulatory protein or other adjuvant, or can encode a combination of immunostimulatory proteins or other therapeutic proteins. The immunostimulatory protein can be a STING protein, such as one that comprises a gain-of-function mutation, or one that is a chimeric STING protein. The bacteria can comprise a plasmid that encodes a combination of therapeutic products. The immunostimulatory protein(s) and/or other therapeutic proteins can be encoded on the plasmid as part of a polycistronic sequence, with the antigen under the control of a prokaryotic promoter recognized by the bacteria; or the immunostimulatory protein(s) and/or other therapeutic proteins can be encoded on the plasmid under control of a eukaryotic promoter recognized by the eukaryotic host. The prokaryotic promoter can be a bacterial promoter or a bacterial phage promoter, and the eukaryotic host can be a human.

The immunostimulatory bacteria can comprise mRNA encoding the antigen(s), and any other proteins expressed under control of a prokaryotic promoter, that are produced by culturing the bacterium in vitro. The immunostimulatory bacteria can comprise genome modifications whereby the bacteria lack flagella and produce LPS with pent-acylated, and/or the bacteria can be asd, and/or an adenosine auxotroph, and/or csgD' and/or ansB~.

The bacteria can comprise nucleic acid encoding a TLR8 agonist.

The bacteria can be msbR IpagP' , and/or can lack flagella, and/or can be asd.

The bacteria can be a species or strain of Escherichia, Listeria, or Salmonella. For example, the bacteria can be Salmonella typhimurium strains, and the unmodified Salmonella strain is a wild-type strain, or is an attenuated strain. The immunostimulatory bacteria can be derived from strain AST- 100 (VNP20009 or YS1646), or from strain ATCC 14028, or from a strain having all of the identifying characteristics of strain ATCC 14028.

As described herein, the immunostimulatory bacteria can contain one or more genome modification that are one or more of a deletion, insertion, disruption, and other modification in a gene, whereby the product encoded by the gene is not produced or, as produced, is inactive.

Also provided herein are pharmaceutical compositions, comprising any of the immunostimulatory bacteria described or provided herein, in a pharmaceutically acceptable vehicle. The pharmaceutical composition can be formulated as a vaccine, for example, as a liquid, a powder, or a tablet. Also provided are methods and uses of the bacteria or pharmaceutical compositions for treating or prevent (reducing the risk of developing) a disease or condition or infection or cancer, as well as uses of the bacteria for delivering RNA, such as mRNA, and methods of delivering the RNA to a subject, comprising administering the bacteria herein.

Also provided are bacteria comprising a plasmid encoding a product or products, where the product(s) is/are a therapeutic product(s), and the plasmid in the bacterium encodes the product(s) to produce mRNA that is not translated by the bacterium.

The bacteria can be attenuated, or rendered of low toxicity or non-toxic, by virtue of the modifications described herein. Exemplary of bacteria are species of Salmonella, such as a Salmonella typhimurium strain. The immunostimulatory bacteria provided herein include those that endogenously encode and express, or are modified to encode and express, a gene encoding resistance to complement killing rck), such as a Salmonella rck gene. For example, therapeutic E. coll are modified to encode rck so that they can be administered systemically.

Protocols, methods, and uses of the therapeutics provided herein

Provided are immunostimulatory bacteria that, upon administration, convert an immune desert or immune excluded tumor into a hot tumor, where: the bacterium comprises genome modifications whereby the bacterium has attenuated TLR2, or attenuated TLR2 and/or TLR4 and/or TLR5 activity, whereby the bacterium, upon administration to a subject, does not produce or produces less of an inflammatory response in the subject to thereby have lower toxicity and higher colonization of tumors compared to a bacterium that does not comprise the genome modifications; the bacterium comprises a plasmid that encodes an immunostimulatory protein that a type I interferon (IFN) or encodes two or more type I interferons (IFNs) or encodes one or more type I interferon (IFN) and another immunostimulatory protein and/or a tumor-associated antigen; and a hot tumor is responsive to immunotherapy or is more responsive than prior to treatment with the immunostimulatory bacterium.

Exemplary of such immunostimulatory bacteria is a strain designated as \6 6Aa cH S LGIApagP AansB AcsgD b-Apurl, or

Y S I 646Azsz// EPEGIApagP/AansB/AcsgD/F-Apurl/AthyA, or YS1646Aasd/AFLG/ApagP/AcmsB/AcsgD/ApurI, or other strains that are pagP /msbB~ , lack curli fimbriae, and are adenosine auxotrophs.

The plasmids in these bacteria encode immunostimulatory bacterium, such as a type I interferon (IFN), such as an IFN-a and/or an IFN-b. The nucleic acid encoding the IFN or IFNs is operatively linked to eukaryotic regulatory sequences. Exemplary promoters and regulatory sequences are well-known to those of skill in the art, such as, where the regulatory sequences comprise an RNA polymerase type II promoter, such as an inducible or constative promoter, and optionally an enhancer, such as where the promoter and enhancer are of viral origin, which is optionally a cytomegalovirus promoter and/or enhancer. The immunostimulatory bacteria can encode at least two immunostimulatory proteins. They can be encoded so that they are transcribed as a polycistronic message that, upon translation results, in the at least to proteins, such as by including a 2A protein or other such regulatory protein or sequence that results in the at least two proteins when transcribed by eukaryotic ribosomes. Where the immunostimulatory bacterium comprises a plasmid that encodes a type I interferon (IFN) or a plurality thereof Exemplary of type I interferons are those having a sequence of amino acids set forth in SEQ ID NOs:550, 552, 549 and 551, and allelic variants or other variants thereof having at least 95% or 98% sequence identity and that have interferon activity. Exemplary sequences are set forth in SEQ ID NOs:549 and 551, which set forth the nucleotide sequence of human IFNa2 and human IFN-b, respectively. SEQ ID NQs:550 and 552 set forth the amino acid sequence of human IFNa2 and human IFN-b, respectively.

The methods, therapeutics, protocols and uses, and immunostimulatory bacterium described herein, including above, can comprise a plasmid that comprises the sequence of nucleotides set forth as SEQ ID NOs:502-545 and degenerate sequences thereof or comprising a portion thereof that comprises nucleic acid encoding the immunostimulatory protein(s), eukaryotic transcription and/or translational regulatory sequences, or sequences having at least 95% sequence identity with the coding portions and regulatory regions of SEQ ID NOs:502-545.

Provided are methods for assessing whether a treatment with a delivery vehicle that is targeted to or phagocytosed by macrophages and that encodes a therapeutic product or products will be effective for treating a tumor in a subject. These methods include identifying proliferating macrophages in a tumor biopsy or tumor sample. The delivery vehicle encodes nucleic acid that is transcribed in the nucleus of a macrophage and is not integrated into a chromosome in the genome (is non-integrating). Delivery vehicles include immunostimulatory bacteria, such any of those provided herein that infect or are phagocytosed by macrophages. It is shown herein that expression of encoded payloads occurs in proliferating macrophages. The proliferating macrophages can be identified by any method known in the art, including methods described herein, where proliferating macrophages are identified in the biopsy by any of the following markers:

Tumor gene expression of G2M module (>14 genes of the set), Stathminl (STMN1);

Biopsy surface markers: CD68 + KI67 and/or PCNA, MERTK;

SPP1 in some tumor types: lung, gastric; and/or

C1QC in some tumor types: colon and breast.

For example, the proliferating macrophages have a hybrid SPP1+ and C1QC+ (expression of both SPP1 and C1QC) macrophage phenotype, and exhibit enhanced phagocytic and proliferating properties.

Also provided are methods for rendering a tumor responsive to immunotherapy, comprising administering a therapeutic that converts macrophages to the M1/M2 hybrid phenotype, wherein the macrophage have been identified as proliferating macrophage. Delivery vehicles and therapeutics that effect such conversion are described throughout the disclosure herein. The resulting macrophages that have a hybrid M1/M2 phenotype have a hybrid SPP1+ and C1QC+ (expression of both SPP1 and C1QC) macrophage phenotype, whereby the macrophage have enhanced phagocytic and proliferating properties.

Provided are plasmids that comprise the sequence of nucleotides set forth in SEQ ID NO:501, or degenerative codons thereof in the protein encoding regions, or a sequence having at least 75%, 80%, 90%, 95%, 96%, 97%, 98%, 99%, or more sequence identity to the sequence set forth in SEQ ID NO:501, where the plasmid encodes a protein that is an IL-15/IL-15R alpha chain complex or a protein having at least 95% sequence identity thereto; and encodes a chimeric STING that constitutively induces type 1 interferon activity and has lower NF-KB signaling activity compared to human STING, and encoding a protein having the activity of IL- 15 receptor complex and a protein that is a STING protein that has constitutive activity and lower NF-KB signaling activity compared to human STING. Provided are immunostimulatory bacteria that comprise the plasmid, including those that have the phenotype YS I646Ac/.sw AFLG/ ApagPl SansBI ScsgD/ -Spur! or YS 1646A< .st/ ^/ AFLG/ pagP! SansBI ScsgD, where YS1646 is /AmsbBISpurl,' and F- pnrl denotes strains in which purl is deleted. The bacteria additionally include those that are thyA~ as described herein. Genome modifications in all embodiments comprise those that render a product or locus inactive, such as by insertion, deletion, transposition, and/or any other change such the encoded active product is not produced.

Provided are immunostimulatory bacteria that comprise a plasmid encoding an IL- 15, such as IL-15/IL-15R alpha chain complex, and encoded a constitutive STING, wherein the plasmid encoding the IL-15/IL-15R alpha chain complex and constitutive STING comprises the sequence set forth in SEQ ID NO:501 or a portion thereof encoding the IL- 15 and eSTING or degenerates thereof or variants thereof having at least 95% sequence identity to the portion or SEQ ID NO: 501 or to degenerates thereof. The immunostimulatory bacterium can be a Salmonella strain that is designated YS I646zlu.sc// AP G/ApagP AansB Acsgl) 1--Apurl or a related strain that has genome modifications that render the strain msbB~!pagP~, lacking flagella, csgD~, and adenosine auxotrophs, and other optional genome modifications as described throughout the disclosure herein.

Provided are protocols or regimens for treating cancers in which PD-1 expression on macrophages in a subject is first suppressed, and then the therapeutic, such as an immunostimulatory bacteria provided herein that converts an immune desert or a T-cell excluded tumor or other tumor that is not responsive to immunotherapy into a hot tumor, responsive to immunotherapy. Following treatment with the therapeutic, the subject can then be treated with an immunotherapy, such as an anti-PD-Ll therapy. An exemplary protocol comprises: pre-treating a subject to be treated with the delivery vehicle with an agent that suppresses PD-1 expression on macrophages to thereby promote the phagocytic capacity of the macrophage, wherein the delivery vehicle comprises nucleic acid encoding an anti-cancer product and is a vehicle that targets or can be phagocytosed by phagocytic macrophages; and then, after a pre-determined time sufficient for the PD-1 expression to be suppressed or reduced, administering the delivery vehicle. The pre-determined time can be at least 4 hours, 5 hours, 6 hours, 7 hours, 8 hours, 9 hours, 10 hours, 12 hours, 14 hours, 16 hours, 18 hours, 20 hours, 22 hours, 24 hours, 1 day to 2 days, up to 72 hours, generally between 4 hours and 48 hours, such as 8 hours to 12 hours, or 4 hours to 24 hours, or 12 hours to 48 hours. For pre-determined time periods as used throughout the disclosure, determination of the time period can be by the skilled person, and can depend upon the particular agents and delivery vehicle administered, as well as other parameters and factors, including the subject. Further immunotherapy, such as anti-PD-Ll therapy also can be administered.

Provided are cancer treatment protocols or regimens that comprise pre-treating a subject to be treated with the delivery vehicle with an anti-PD-1 agent, whereby PD- 1 expression on the macrophages is suppressed to thereby promote phagocytic capacity of the macrophages, wherein the delivery vehicle comprises nucleic acid encoding an anti-cancer product and is a vehicle that targets or can be phagocytosed by phagocytic macrophages; and then administering the delivery vehicle. The delivery vehicle is administered after a sufficient time for the PD-1 expression to be suppressed or reduced, such as at least 4 hours, 5 hours, 6 hours, 7 hours, 8 hours, 9 hours, 10 hours, 12 hours, 14 hours, 16 hours, 18 hours, 20 hours, 22 hours, 24 hours, 1 day to 2 days, up to 72 hours, or such as between 8 hours and 48 hours, or 4 hours and 12 hours, or 8 hours and 24 hours. Generally the delivery vehicle, such as an immunostimulatory bacterium is administered within 24 to 48 hours after the administration of the anti-PD-1 therapy. The protocol can include a further step of treating with an immunotherapy agent such as an immune checkpoint inhibitor. The immunotherapy is administered after the delivery agent, generally after a predetermined time sufficient for the tumor to be susceptible to immunotherapy. For example, the protocol can include, after the anti-cancer product encoded in the delivery vehicle is expressed, whereby PD-L1 is expressed on the macrophages, then administering an anti-PD-Ll agent. The pre-determined time for the immunotherapy is at least 4 hours, 5 hours, 6 hours, 7 hours, 8 hours, 9 hours, 10 hours, 12 hours, 14 hours, 16 hours, 18 hours, 20 hours, 22 hours, 24 hours, 1 day to 2 days, up to 72 hours, or such as between 8 hours and 48 hours, or 4 hours and 12 hours, or 8 hours and 24 hours. In these protocols or regimens, the anti-PD-Ll agent can be an anti-PD- Ll antibody, such as an anti-PD-Ll antibody or other antagonist.

The delivery agent in the protocols and regimens can be one that converts macrophages that phagocytose the delivery vehicle into macrophages with the M1/M2 hybrid phenotype as described herein, such as any immunostimulatory bacterium or other therapeutic described throughout the disclosure herein as having this property, Exemplary of such a delivery agent is the bacterium designated “test strain 4” in the table in Example 39, and related strains. The protocol or regimen comprises: administering an PD-1 antibody, then administering the bacterium, and then administering an immunotherapy, such as an anti-PD-Ll antibody.

Methods of treating a subject who has an immune desert or T-cell excluded tumor, comprising: first administering an agent that suppresses PD-1 expression on macrophages, whereby phagocytic capacity of the macrophages is increased relative to before treatment; and then after a pre-determined time, administering a delivery vehicle that comprises nucleic acid encoding an anti-cancer product, where the delivery vehicle targets or accumulates in phagocytic macrophages; then administering the delivery vehicle. The pre-determined time can be at least 4 hours, 5 hours, 6 hours, 7 hours, 8 hours, 9 hours, 10 hours, 12 hours, 14 hours, 16 hours, 18 hours, 20 hours, 22 hours, 24 hours, 1 day to 2 days, up to 72 hours, or such as between 8 hours and 48 hours, or 4 hours and 12 hours, or 8 hours and 24 hours. The pre-determined time is sufficient for suppression of PD-1 expression on the macrophages, such as, for example, at least 4 hours, 5 hours, 6 hours, 7 hours, 8 hours, 9 hours, 10 hours, 12 hours, 14 hours, 16 hours, 18 hours, 20 hours, 22 hours, 24 hours, 1 day to 2 days, up to 72 hours, or such as between 8 hours and 48 hours, or 4 hours and 12 hours, or 8 hours and 24 hours, or 12 hours and 48 hours, or other such time period as determined by the skilled person. The methods an further include, after a second pre-determined time period, such as a time period recited above, administering immunotherapy, such as a checkpoint inhibitor, such as an anti-PD-Ll antibody or other inhibitor, to the subject.

In accord with the protocols, regimens, methods, and uses described above, the delivery vehicle can be an immunostimulatory bacterium that encodes a immunostimulatory protein, such as any described herein, such as a Salmonella species that has genome modifications whereby the bacterium lacks flagella, is an adenosine auxotroph, has penta-acylated LPS, and optionally lacks curli fimbriae, and/or is asd~, and optionally other genome modifications that reduce toxicity /inflammatory responses to the bacterium, and/or promotes accumulation/targeting in phagocytic macrophage. The immunostimulatory bacteria can encode one or more immunostimulatory protein(s), and optionally a tumor- associated antigen under control of a eukaryotic regulatory signals. Combinations of immunostimulatory protein include those described above and throughout the disclosure herein. Exemplary of immunostimulatory protein(s) are one or more of a cytosolic DNA/RNA sensor, a cytokine, and/or a tumor-associated antigen, such as combinations set forth above and below. Exemplary of immunostimulatory proteins are cytosolic DNA/RNA sensors, such as a modified STING protein that constitutively induces type I interferon (IFN) in macrophages (referred to as eSTING in disclosure herein), a cytokine that is IL-15 or IL-15/IL-15R alpha chain complex and/or a type I interferon (IFN) or other cytokine with similar properties to IL- 15 and IL-15/IL-15R alpha chain complex. An exemplary bacterium is the bacterium is Y S I 646 < .V<7/ ^' G/ApagP/AansB/AcsgD/F-Apurl containing a plasmid encoding IL- 15/IL-15R alpha chain complex and the chimeric STING with the CTT from Tasmanian devil and the replacement N154S/R284G, such as the strain designated CRST-2000 (see Example 39) or a derivative thereof that has additional genome modifications and related strains, is/are exemplary of immunostimulatory bacteria for use in the methods, uses, protocols and other embodiments provided herein.

Provided are immunostimulatory bacteria, such as the bacteria with genome modifications as described herein, where the immunostimulatory bacterium is for use for or for use in a method for converting an immune desert or T-cell excluded tumor into hot tumors. In accord with these uses and methods, the subject to be treated is identified as having an immune desert or T-cell excluded tumor. The bacterium is administered and, following treatment, the tumors are susceptible to treatment with an immune checkpoint inhibitor or other immunotherapy was not effective for treating the tumor prior to treatment with the immunostimulatory bacterium. The immunostimulatory bacteria are any described herein, such as those that result in the M1/M2 hybrid phenotype, such as those the encode the IL-15/IL-15R alpha chain complex and eSTING that results in constitutive expression of type I interferon (IFN), and also the immunostimulatory bacteria that encode one or more of an IFNa and/or IFNb, such as the bacteria containing plasmids (or portions thereof) described in Example 57, such as a bacterium that comprises a plasmid that contains all or a part of a nucleic acid molecule with the nucleotide sequence set forth in SEQ ID NOs. 502- 545 or containing degenerate codons thereof. The plasmids and portions thereof an IFNa and/or an IFNb or variant thereof that has IFNa or IFNb activity. The immunostimulatory proteins are encoded and expressed under control of a eukaryotic promoter, and optionally other regulatory sequences. Isolated macrophages for cell therapy

Provided are isolated macrophages, comprising a therapeutic that, when introduced in the macrophage, results in a M1/M2 hybrid macrophage phenotype The therapeutic is introduced into the macrophages in vitro or ex vivo. Following culturing or treatment or formulation of the macrophages comprising the therapeutic, the resulting compositions containing the macrophages can be administered to a subject in need of treatment with the macrophages, such as a subject with cancer, such as a cancer comprising tumors referred to as T-cell excluded tumors or desert tumors or cold tumors to thereby convert the tumors into hot tumors that are susceptible to immunotherapy. The macrophages can be allogeneic or autologous to the subject to be treated. The therapeutic can be one, such as immunostimulatory bacteria described herein, that induces a hybrid M1/M2 phenotype, whereby the macrophage can phagocytose apoptotic tumor cells, induce constitutive type I IFN to recruit and prime tumor antigen-specific CD8+ T-cells, and thereby induce durable anti-tumor immunity. The macrophages can be isolated from a subject or previously obtained and cultured, and optionally genetically modified; the therapeutic is introduced into the macrophages, which generally are cultured. The macrophages can be formulated for storage prior to use or for administration. The therapeutic includes any described herein or identified as converting macrophages into an M1/M2 phenotype. Therapeutics include immunostimulatory bacteria that encode payloads for treatment of cancers or for other applications, such as for RNA delivery, as vaccines, or treatment of diseases, disorders, and conditions. Included are immunostimulatory bacteria that encode two or more complementary immunostimulatory proteins. Exemplary of the immunostimulatory bacteria are those that comprise the phenotype YS1646/ AFLG/ pagPi csgD or YS I 646Aav// AFLG/ pagP/ ansBI iXcsg!) or Y S 1646 asd! AFLG / /SpagP/ tXcsgl), or other such phenotype resulting in reduced or eliminated TLR2/4/5 responses, such that the bacteria have reduced inflammatory properties compared to VNP20009, and primarily or solely infect phagocytic cells, such as macrophages. Exemplary of therapeutics are those that are delivery vehicles that comprise a nucleic acid molecule of SEQ ID NO:501, or a sequence having at least 90% or 95% or 97% or 98% or 99% or more sequence identity to SEQ ID NO:501, or a nucleic acid molecule comprising one or more degenerate codons to either SEQ ID NO: 501 or the sequence having at least 90% or 95% or 97% or 98% or 99% or more sequence identity to SEQ ID NO: 501. Hence provided are uses of the therapeutics and macrophages for treatment of cancer, where the macrophages following introduction of the therapeutic are introduced into a subject with cancer or used for treatment of cancer, such as in a subject that has a T-cell excluded (or immune desert or cold tumor) or has or does not respond to immunotherapy, such as anti-PDl immunotherapy. Provided are methods for converting an immune desert tumor or T-cell excluded tumor or cold tumor into a hot tumor, comprising administering isolated macrophages that contain the therapeutic.

Also provided, as described herein, and as set forth in the claims, are delivery vehicles, cells, pharmaceutical compositions, methods, uses, and treatments of cancer, particularly in humans. Also provided are companion diagnostics and methods for selection of subjects for treatment, and methods for monitoring treatment. These are described below, and also, in the claims, which are incorporated in their entirety into this section.

BRIEF DESCRIPTION OF THE FIGURES

Figures 1A-1C depict the inserts in plasmids pATI-1.75 and pATI-1.76 (Fig. 1A and IB, respectively). Figure 1C depicts that the Shine-Dalgarno sequence is replaced with a Kozak sequence, for translation in eukaryotic cells, such as myeloid cells.

Figure 2 depicts the alignment of wild-type human STING (SEQ ID NO:306) and Tasmanian devil STING (SEQ ID NO:349) proteins.

Figure 3 depicts the alignment of wild-type human STING (SEQ ID NO:306) and marmoset STING (SEQ ID NO:359) proteins.

Figure 4 depicts the alignment of wild-type human STING (SEQ ID NO:306) and cattle STING (SEQ ID NO:360) proteins.

Figure 5 depicts the alignment of wild-type human STING (SEQ ID NO:306) and cat STING (SEQ ID NO:356) proteins.

Figure 6 depicts the alignment of wild-type human STING (SEQ ID NO:306) and ostrich STING (SEQ ID NO:361) proteins.

Figure 7 depicts the alignment of wild-type human STING (SEQ ID NO:306) and crested ibis STING (SEQ ID NO:362) proteins.

Figure 8 depicts the alignment of wild-type human STING (SEQ ID NO:306) and coelacanth STING (SEQ ID NO:363) proteins. Figure 9 depicts the alignment of wild-type human STING (SEQ ID NO:306) and zebrafish STING (SEQ ID NO:348) proteins.

Figure 10 depicts the alignment of wild-type human STING (SEQ ID NO:305) and boar STING (SEQ ID NO:365) proteins.

Figure 11 depicts the alignment of wild-type human (SEQ ID NO:305) and bat STING (SEQ ID NO:366) proteins.

Figure 12 depicts the alignment of wild-type human (SEQ ID NO:305) and manatee STING (SEQ ID NO:367) proteins.

Figure 13 depicts the alignment of wild-type human (SEQ ID NO:305) and ghost shark STING (SEQ ID NO:368) proteins.

Figure 14 depicts the alignment of wild-type human (SEQ ID NO:305) and mouse STING (SEQ ID NO: 369) proteins.

Figure 15 is adapted from Roszer et al., ((2018) Cells 7(8): 103). Roszer et al. presents a scheme summarizing signals that impede or facilitate cell cycle entry and progression in macrophages.

Figure 16 depicts priming and activation of tumor-associate antigen (TAA)- specific CD8+ T-cells and induction of anti-tumor immunity by the exemplary immunostimulatory bacterium, designated as STACT, that encodes IL-15/IL-15R alpha chain complex + eSTING that constitutively induces type I IFN, such as an eSTING.

Figures 17A-F. FIG. 17A depicts a schematic from Anfray et al. Cells (2019) which lists molecular pathways previously associated with tumor-associated macrophage (TAM) phenotypes. FIG. 17B depicts a cartoon of an Ml macrophage (top) and M2 macrophage (bottom) and resulting macrophage with an M2 hybrid phenotype, described herein. FIG. 17C describes the macrophage markers that comprise the M1/M2 hybrid phenotype described herein and the results upon infection of tumor-resident macrophage by the immunostimulatory bacterium described herein and expression of the encoded payload. FIG. 17D is from Roszer, T. ((2015) Mediators Inflamm 2015:816460) which represents examples of M2 activation and markers associated with the distinct activation phenotypes of Ml and M2, and not a hybrid M1/M2. FIG. 17E depicts the STACT mechanism of action. FIG. 17F is adapted from a figure depicting the Cancer-Immunity Cycle in Chen and Mellman ((2013) Immunity 39(1): 1-10) and depicts STACT IL-15plex + eSTING as a Comprehensive Immunotherapy. Figure 18 depicts SPP1 expression in tumor tissue compared to normal tissue in a variety of tumor types and tissues. Color of p-value text and shape surrounding p- value represents direction; gray p-value surrounded by oval = lower in tumor (Neg); light gray p-value surrounded by rectangle, higher in tumor (Pos); and very light gray p-value with no shape around p-value, no significance (NS). The cancer types are ordered by the median expression of the tumor samples (left to right corresponding to high to low).

Figure 19 depicts C1QC expression in tumor tissue compared to normal tissue in a variety of tumor types and tissues. Color of p-value text and shape surrounding p- value represents direction; gray p-value surrounded by oval = lower in tumor (Neg); light gray p-value surrounded by rectangle, higher in tumor (Pos); and very light gray p-value with no shape around p-value, no significance (NS). The cancer types are ordered by the median expression of the tumor samples (left to right corresponding to high to low).

Figure 20 depicts a Kaplan-Meier (KM) plot of a bioinformatics analysis of groups with high or low C1QC expression in consensus molecular subtype 1 (CMS1) colorectal cancer.

Figure 21 depicts a forest plot showing the results for Gene expression of 23 tumor types from The Cancer Genome Atlas (abbreviated TCGA). The Cox proportional hazard regression model was calculated, to test for the association between prognosis and expression of SPP1 as a continuous variable. The shape of the dot represents the p-value; a triangle represents p < 0.05. The dot is the hazard ratio for each comparison and the black lines are the 95% confidence intervals.

Figure 22 depicts a forest plot showing the results for Gene expression of 23 tumor types from The Cancer Genome Atlas (TCGA). The Cox proportional hazard regression model was calculated, to test for the association between prognosis and expression of C1QC as a continuous variable. The shape of the dot represents the p- value; a triangle represents p < 0.05. The dot is the hazard ratio for each comparison and the black lines are the 95% confidence intervals.

Figures 23A-B depict the quantity of proliferating macrophages across histologies based on analysis was performed on tumor types using published scRNA- seq datasets. The proportions are depicted in the graphs as percentages (0.05 = 5%) and calculated using proportion of CD68+ cells of total CD45+ cells (FIG. 23A), and proportion of G2/M score+ of total CD68+ macrophages (FIG. 23B) from tumor tissue samples of all patients in each dataset. Boxes represent median ± interquartile range and whiskers ±1.5 x interquartile range. Outliers are represented by black dots.

Figure 24 depicts a box and whisker plots generated from the apoptosis module expression data from The Cancer Genome Atlas (abbreviated as TCGA; 161 genes associated with apoptosis), compared to cancer types and ordered by median expression.

Figures 25A-F depict box plots of the G2M score (FIGs. 25A-C) and STMN1 expression (FIGs. 25D-F) in non-proliferating and proliferating macrophages in lung cancer (A, D); breast cancer (B, E); and Colon cancer (not stratified into CMS subtypes) (C, F).

Figure 26 depicts association plots that show association of SPP1 expression with alterations in major cancer pathways, where the y-axis represents the -log 10(p- value), where the higher the dot corresponds to a lower p-value. The different dots represent tumor type.

Figure 27 depicts association plots that show association of C 1 QC with alterations in major cancer pathways, where the y-axis represents the -loglO(p-value), where the higher the dot corresponds to a lower p-value. The different dots represent tumor type.

Figure 28 depicts a box plot of CD68 expression in breast cancer patients at baseline, or following either 2 cycles of epirubicin and docetaxol (C2, chemo only), or 4 cycle of chemotherapy + surgery and bevacizumab (C4, sur±chemo±bev).

Figure 29 depicts a box plot of CD68 expression in breast cancer patients at baseline, or following either 2 cycles of epirubicin and docetaxol (C2, chemo only), or 4 cycle of chemotherapy ± surgery and bevacizumab (C4, sur±chemo±bev).

Figure 30 depicts pHRodo® reagent, a dye that is fluorescent only in acidic cellular components, and CellTrace® reagent labeling of

YS 1646Aasr// AFLG/ kpagP/ kansBI kcsgD containing the plasmid encoding IL- 15 receptor complex and chimeric STING, where the plasmid has the sequence set forth in SEQ ID NO:501, internalized by THP-1 differentiated into MO macrophages over time. pHRodo® and CellTrace®reagent labeling is shown after bactofection at MOI 1, 5, 20 and 40 at 90 minutes (top panel) and 48 hours (bottom panel).

Figures 31A-F depict the uptake of YS1646Aast//AFLG/Agag77Aany6/AcsgD containing the plasmid encoding IL- 15 receptor complex and chimeric STING, where the plasmid has the sequence set forth in SEQ ID NO:501 , internally by M0, Ml, and

- 109 -

RECTIFIED SHEET (RULE 91 ) ISA/EP M2 macrophages after bactofection at MOI 1, 5, 20 and 40. The uptake is depicted in MO macrophages as measured by pHRodo® reagent (FIG. 31A) and CellTrace® reagent (FIG. 31D); in Ml macrophages as measured by pHRodo® reagent (FIG. 31B) and CellTrace® reagent (FIG. 31E); and in M2 macrophages as measured by pHRodo® (FIG. 31C) and CellTrace® reagent (FIG. 31F).

Figures 32A and 32B depict STING reporter luciferase activity in wild-type MO THP-1 derived macrophages; THP-1 cells that were not bactofected and THP-1 cells to which cGAMP was added also is shown. FIG. 32A depicts STING reporter luciferase activity in wild-type MO THP-1 derived macrophages at MOI 1, 5, 20 and 40 measured at 90 minutes, 3 hours, 24 hours, and 48 hours after bactofection. FIG. 32B depicts STING reporter luciferase activity in wild-type M0 THP-1 derived macrophages at MOI 1, 5, 20 and 40 measured 48 hours after bactofection.

Figure 33 depicts the cell cycle-dependent internalization of STACT (as referred to in Example 58, YS 1646A< .sc// Ah LG/ EpagPI EansBI csgl) containing the plasmid encoding IL-15 receptor complex and chimeric STING, where the plasmid has the sequence set forth in SEQ ID NO:501). The top left panel shows the quadrant of the graph that corresponds to cell cycle phase. The bottom left panel depicts the cell cycle phase of non-stained cells. The top right panel depicts the cell cycle phase for M0 cells that were not bactofected (left), bactofected at MOI 20 and 40 (middle), and cells bactofected with a bacterium comprising asd but no plasmid as a control (right). The middle right panel depicts the cell cycle phase for Ml cells that were not bactofected (left), bactofected at MOI 20 and 40 (middle), and cells with ASD (right). The bottom right panel depicts the cell cycle phase for M2 cells that were not bactofected (left), bactofected at MOI 20 and 40 (middle), and cells with ASD (right).

Figure 34 is a schematic that depicts the high level of adenosine and immune exclusion that is triggered by hypoxia in tumors.

Figures 35A and 35B depict tumors with high CD73 and CD39 expression. FIG. 35A depicts a box and whisker plot showing the median expression of CD73 for each tumor indication from the Atlas (abbrev. TCGA). FIG. 35B is a scatterplot depicting expression of the ENTPD-1 gene (CD39) verses expression of the NT5E gene (CD73) for tumor indications in the TCGA; triangles represent top indications, and the tumor indications are shown in gradations according to Kummel myeloid signatures. Figures 36A and 36B depict Krummel T cell, myeloid, and stromal cell signatures across solid tumor indications in TCGA. Triangles represent top indications, and the tumor indications are shown in gradations according to NT5E (CD73) expression. FIG. 36A is a scatterplot depicting Krummel T cell verses myeloid cell signatures. FIG. 36B is a scatterplot depicting Krummel stromal cell verses myeloid cell signatures.

Figures 37A-F depict myeloid cell markers from the TCGA across the C1-C6 immune subtypes (wound healing (Cl), IFN-y dominant (C2), inflammatory (C3), lymphocyte depleted (C4), immunologically quiet (C5), and TGF-P dominant (C6)) defined in Thorsson et al. (2018) Immunity 48(4):812-830. The key characteristics of each immune subtype is shown in FIG. 37A. Overall survival in years is shown for each immune subtype in FIG. 37B. In FIGs. 37C-F, box and whisker plots depict Krummel T cell signatures for each subtype from TCGA (FIG. 37C); Krummel myeloid cell signatures for each subtype from TCGA (FIG. 37D); MRC1 expression for each subtype from TCGA (FIG. 37E); and CD 163 expression for each subtype from TCGA (FIG. 37F).

Figure 38 depicts tumor indications with high myeloid cell signatures, T cell signatures, hypoxia markers, adenosine markers, TGFbeta and antigen presentation in TCGA.

Figure 39 depicts metastatic cancer indications with high myeloid cell signatures, T cell signatures, hypoxia markers, adenosine markers, TGFbeta and antigen presentation in MET500 dataset.

Figures 40A-C depict syngeneic mouse models with high T cell, myeloid and stromal infiltration and high adenosine from the TISMO database. FIG. 40A is a plot depicting a rank-based scoring for a set of genes within fibroblast signatures verses a set of genes within myeloid cell signatures for syngeneic mouse models. The models are shown in gradations according to the rank-based scoring for a set of genes within T cell signatures. FIG. 40B is a plot depicting a rank-based scoring for a set of genes within T cell signatures verses a set of genes within myeloid cell signatures for syngeneic mouse models. The models are shown in gradations according to Cd8a expression. FIG. 40C is a plot depicting the expression c NT5E (CD73) verses CD68, a marker for high myeloid content, for syngeneic mouse models. The models are shown in gradations according to Stingl expression.

- I l l - Figure 41 depicts box and whisker plots generated from SPP1 expression across TCGA, and compared to cancer types ordered by median expression. The COAD and READ colorectal cancers were separated into their CMS subtypes and plotted.

Figure 42 depicts box and whisker plots generated from C1QC expression across TCGA, and compared to cancer types ordered by median expression. The COAD and READ colorectal cancers were separated into their CMS subtypes and plotted.

DETAILED DESCRIPTION

OUTLINE

A. DEFINITIONS

B. OVERVIEW OF IMMUNO STIMULATORY BACTERIA FOR CANCER THERAPY

1. Bacterial Cancer Immunotherapy

2. Prior Therapies that Target the Tumor Microenvironment a. Limitations of Autologous T-Cell Therapies b. Viral Vaccine Platforms c. Bacterial Cancer Therapies i. Listeria ii. Salmonella Species iii. VNP20009 (YS1456) iv. Wild-Type Strains

3. Limitations of Existing Bacterial Cancer Immunotherapies

4. Therapeutics That Induce a Hybrid M1/M2 Anti-tumor Phenotype in Tumor-Resident Macrophages

C. MODIFICATIONS AND ENHANCEMENTS OF IMMUNOSTIMULATORY BACTERIA TO INCREASE THERAPEUTIC INDEX AND TO INCREASE ACCUMULATION IN TUMOR-RESIDENT MYELOID CELLS

1. Deletions in Genes in the LPS Biosynthetic Pathway a. msbB Deletion b. pagP Deletion or Inactivation

2. Nutrient Auxotrophy a. purl Deletion/Disruption b. Adenosine Auxotrophy c. Thymidine Auxotrophy

3. Plasmid Maintenance and Delivery a. asd Deletion b. endA Deletion/Disruption 4. Flagelhn Knockout Strains

5. Engineering Bacteria to Promote Adaptive Immunity and Enhance T-Cell Function

L-Asparaginase II (ansB) Deletion/Disruption

6. Deletions/Disruptions in Salmonella Genes Required to Produce Curli Fimbriae csgD Deletion

7. Improving Resistance to Complement

Rck Expression

8. Deletions of Genes Required for Lipoprotein Expression in Salmonella and Other Gram-Negative Bacteria

9. Robust Immunostimulatory Bacteria Whose Genomes are Optimized for Anti-Tumor Therapy, and that Encode Therapeutic Products, Including a Plurality Thereof

10. Vaccines and Bacteria that Deliver RNA, including mRNA and other Forms of RNA, for Expression in a Eukaryotic Host

11. Bacterial Vaccines Against Particular Antigens, Including from Pathogens, and also from Tumors, for use as AntiPathogen Treatments and Vaccines, and For Anti-Cancer Treatment and/or Prevention

12. Conversion of M2 Phenotype Macrophages into Ml and Ml-Like Phenotype Macrophages

D. IMMUNOSTIMULATORY BACTERIA WITH ENHANCED THERAPEUTIC INDEX ENCODING GENETIC PAYLOADS THAT STIMULATE THE IMMUNE RESPONSE IN THE TUMOR MICROENVIRONMENT

1. Immunostimulatory Proteins a. Cytokines and Chemokines b. Co-Stimulatory Molecules

2. Constitutively Active Proteins that Stimulate the Immune Response and/or Type I IFN, Non-Human STING Proteins, STING Chimeras, and Modified Forms a. Constitutive STING Expression and Gain-of- Function Mutations b. Constitutive IRF3 Expression and Gain-of-Function Mutations c. Non-Human STING Proteins, and Variants Thereof with Increased or Constitutive Activity, and STING Chimeras, and Variants Thereof with Increased or Constitutive Activity d. Other Gene Products that Act as Cytosolic DNA/RNA Sensors and Constitutive Variants Thereof i. RIG-I ii. MDA5/IFIH1 iii. IRF7 e. Other Type I IFN Regulatory Proteins

3. Antibodies and Antibody Fragments a. TGF-p b. Bispecific scFvs and T-Cell Engagers c. Anti-PD-1, Anti-PD-Ll and Anti-CTLA-4 Antibodies i. Anti-PD-l/Anti-PD-Ll Antibodies ii. Anti-CTLA-4 Antibodies d. Additional Exemplary Checkpoint Targets

4. Combinations of Immunomodulatory Proteins can have Synergistic Effects and/or Complementary Effects

5. Molecules that Activate Prodrugs

6. Immunostimulatory Bacteria that Deliver Combination Therapies

E. IMMUNOSTIMULATORY BACTERIA AS ANTIVIRAL THERAPEUTICS AND AS THERAPEUTICS AGAINST OTHER INFECTIOUS PATHOGENS

F. CONSTRUCTING EXEMPLARY PLASMIDS ENCODING THERAPEUTIC PRODUCTS FOR BACTERIAL DELIVERY

1. Constitutive Promoters for Heterologous Expression of Proteins

2. Multiple Therapeutic Product Expression Cassettes a. Single Promoter Constructs b. Dual/Multiple Promoter Constructs

3. Regulatory Elements a. Post-Transcriptional Regulatory Elements b. Polyadenylation Signal Sequences and Terminators c. Enhancers d. Secretion Signals e. Improving Bacterial Fitness

4. Origin of Replication and Plasmid Copy Number

5. CpG Motifs and CpG Islands

6. Plasmid Maintenance/Selection Components

7. DNA Nuclear Targeting Sequences G. EXEMPLARY BACTERIAL STRAINS AND MECHANISM OF ACTION FOR USE AS VACCINES AND THERAPEUTICS

1. Exemplary Immunostimulatory Bacteria — In Situ Cancer Vaccination Mechanism of Action (MOA)

2. Exemplary Immunostimulatory Bacteria for Peripheral Cancer Vaccination and Mechanism of Action

3. Exemplary Immunostimulatory Bacteria for Pathogen Vaccination and MOA

4. Exemplary Immunostimulatory Bacteria for Treatment of Cancer

H. PHARMACEUTICAL PRODUCTION, COMPOSITIONS, AND FORMULATIONS

1. Manufacturing a. Cell Bank Manufacturing b. Drug Substance Manufacturing c. Drug Product Manufacturing

2. Compositions

3. Formulations a. Liquids, Injectables, Emulsions b. Dried Thermostable Formulations

4. Compositions for Other Routes of Administration

5. Dosages and Administration

6. Packaging and Articles of Manufacture

I. METHODS OF TREATMENT AND USES

1. Diagnostics for Patient Selection for Treatment and for Monitoring Treatment a. Patient Selection b. Diagnostics to Assess or Detect Activity of the Immunostimulatory Bacteria are Indicative of the Effectiveness of Treatment

2. Tumors

3. Administration

4. Monitoring

J. EXAMPLES

A. DEFINITIONS

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as is commonly understood by one of skill in the art to which the invention(s) belong. All patents, patent applications, published applications and publications, GenBank sequences, databases, websites and other published materials referred to throughout the entire disclosure herein, unless noted otherwise, are incorporated by reference in their entirety. In the event that there are a plurality of definitions for terms herein, those in this section prevail. Where reference is made to a URL or other such identifier or address, it is understood that such identifiers can change and particular information on the internet can come and go, but equivalent information can be found by searching the internet. Reference thereto evidences the availability and public dissemination of such information.

As used herein, STACT refers to the S. Typhimurium-Attenuated Cancer Therapy strain. STACT generally refers to the exemplary strain designated YS1646Aasd/AFLG/ApagP/AansB/AcsgD. YS1646, described in the detailed description is msbB~ and purl~ by virtue of genome modifications that disrupt expression of the gene products. STACT includes these modifications, and optionally includes full gene deletions of either or both msbB and purl. The STACT strain in the Examples includes a full deletion of purl. The STACT strains modifications, including genome modifications that eliminate flagella and render the bacteria csgD' in addition to msbB and purl~. The modifications that render the bacterium asd and/or cmsB~ are optional, and are user selected for a particular application or protocol. The STACT strains are exemplary of the immunostimulatory bacteria described and provided herein. These immunostimulatory bacteria, including the exemplary bacteria designated STACT, comprise the genome modifications that delete flagella and biofilm (csgD~) and result in penta-acylated LPS. This includes strains that comprise the phenotype asd/ AFLG/ pagPI msB/ csg!) , with modifications of purl and msbB~ (either by insertions, deletions, transpositions, or complete deletion of each gene) result in advantageous properties discussed and demonstrated throughout the disclosure herein. These properties include, but are not limited to, for example: (1) enhanced, compared to the unmodified parental strain YS1467 (also designated VNP20009), tolerability after IV dosing, (2) tumor-specific enrichment, (3) phagocytosis by tumor-resident antigen-presenting cells (APCs) with a lack of epithelial cell infectivity, (4) provision of multiplexed genetic cargo delivery, and (5) attenuation of bacterial pathways that impair CD8 ⁺ T-cell function. STACT bacteria can encode payloads, such as therapeutic proteins, such as immunostimulatory proteins, such as cytokines, co-stimulatory molecules, cyclic DNA/RNA sensors, particularly those modified to constitutively induce type I interferon (IFN), type I interferon (IFN)s, tumor-associated antigens or other antigens, antibodies, and other such proteins. As used herein, “therapeutic bacteria” are bacteria that effect therapy, such as anti-cancer or anti -tumor therapy, when administered to a subject, such as a human.

As used herein, a therapeutic that is provided herein comprises a delivery vehicle and nucleic acid, such as DNA, that is delivered to cells or tissues, such as tumor-resident immune cells, the tumor microenvironment, and tumor, where it is taken up by cells, such as tumor-resident immune cells and the nucleic acid is expressed in the cells. Generally the nucleic acid encodes one or more immunostimulatory proteins, including proteins that induce type I IFN expression, and the delivery vehicle is designed or formed or formulated so that it does not induce sufficient TLR2 or combinations of one or more of TLR2, TLR4, and TLR5 activity to inhibit type I IFN, such as the that induced by an encoded immunostimulatory protein, such as STING protein.

As used herein, a therapeutic targeted to particular tissues or cells, such as tumors, refers to active targeting in which the therapeutic is directed to tissue or cells, such as by including a protein that binds to a cell surface protein, and also passive targeting in which something accumulates in a cells, such as a cell in which the targeted therapeutic ends up because it cannot be taken up by other cells. A tumor- targeted therapeutic is a therapeutic that ends up in or accumulates in the tumor microenvironment, cells in the tumor microenvironment, and/or the tumor. Generally tumor-targeted therapeutics do not end up in or minimally end up in other cells and tissues and non-tumor loci.

As used herein, “immunostimulatory bacteria” are therapeutic bacteria that, when introduced into a subject, accumulate in immunoprivileged tissues and cells, such as tumors, the tumor microenvironment and tumor-resident immune cells, and replicate and/or express products that are immunostimulatory or that result in immunostimulation. For example, the immunostimulatory bacteria are attenuated in the host by virtue of reduced toxicity or pathogenicity and/or by virtue of encoded products that reduce toxicity or pathogenicity, as the immunostimulatory bacteria cannot replicate and/or express products (or have reduced replication/product expression), except primarily in immunoprivileged environments, such as the tumor microenvironment (TME). Immunostimulatory bacteria provided herein are modified to encode a product or products, and/or to exhibit a trait or property that renders them immunostimulatory. The immunostimulatory bacteria also include genome modifications so that an endogenous product or products is/are not expressed. The bacteria can be said to be deleted in such product(s). Those of skill in the art recognize that genes can be inactivated by deletions, disruptions, including transposition or insertion of transposons, insertions, and any other changes that eliminate the gene product. This can be achieved by insertions, deletions, and/or disruptions, including transpositions or inclusion of transposons. Examples of genes that are inactivated include, for example, msbB, pagP, ansB, gene(s) encoding curli fimbriae, genes encoding flagella whereby the bacterium lacks flagella, and other modifications described herein and/or known to those of skill in the art. Those of skill in the art also understand that corresponding genes in various bacterial species may have different designations. The encoded products, properties and traits in the immunostimulatory bacteria, include, but are not limited to, for example, at least one of: an immunostimulatory protein, such as a cytokine, chemokine, or co-stimulatory molecule; a cytosolic DNA/RNA sensor or gain-of-function or constitutively active variant thereof (e.g., STING, IRF3, IRF7, IRF-8, MDA5, RIG-I); RNAi, such as siRNA (shRNA and microRNA), or CRISPR, that targets, disrupts, or inhibits an immune checkpoint, such as, for example, TREX1, PD-1, CTLA-4 and/or PD-L1; antibodies and fragments thereof, such as an anti-immune checkpoint antibody, an anti-IL-6 antibody, an anti-VEGF antibody, or a TGF-p inhibitory antibody; other antibody constructs, such as bi-specific T-cell engagers (BiTE® antibodies); soluble TGF-P receptors that act as decoys for binding TGF-p, or TGF-P antagonizing polypeptides; and IL-6 binding decoy receptors. Immunostimulatory bacteria also can include a modification that renders the bacteria auxotrophic for a metabolite that is immunosuppressive or that is in an immunosuppressive pathway, such as adenosine.

As used herein, the strain designations VNP20009 (see, e.g., International PCT Application Publication No. WO 99/13053, see, also U.S. Patent No. 6,863,894), YS1646, and 41.2.9 are used interchangeably, and each refers to the strain deposited with the American Type Culture Collection (ATCC) and assigned Accession No. 202165. VNP20009 is a modified attenuated strain of Salmonella typhimurium, which contains deletions or other modifications in msbB and purl, and was generated from wild-type S. typhimurium strain ATCC #14028.

As used herein, the strain designations YS1456 and 8.7 are used interchangeably and each refer to the strain deposited with the American Type Culture Collection (ATCC) and assigned Accession No. 202164 (see, U.S. Patent No. 6,863,894). This strain msbB~ and purP, and is the VNP2009 strain. As used herein, recitation that a bacterium is “derived from” a particular strain means that such strain can serve as a starting material and can be modified to result in the particular bacterium.

As used herein, T cell exhaustion is a state of T cell dysfunction that arises during many chronic infections and cancer. It describes the response of T cells to chronic antigen stimulation in these settings. It is defined by poor effector function, sustained expression of inhibitory receptors, and a transcriptional state distinct from that of functional effector or memory T cells. It is characterized by the stepwise and progressive loss of T-cell functions.

As used herein, a gene module is a set or group of genes with similar expression profiles or by one or more genetic or cellular interactions, such as a set of co-expressed genes to which the same set of transcription factors bind. For example, a G2M score is a set of genes associated with the cell cycle transition from G2 to M, and thus, serves as an indicating or cell proliferation.

As used herein, proliferating macrophage can be identified by some or all of the following: Tumor gene expression of G2M module (>14 genes of the set), and Stathminl (STMN1); and/or

Biopsy surface markers: CD68 + KI67 and/or PCNA, MERTK. Proliferating macrophage can exhibit all of the above markers or a subset thereof. For example, gene expression of the G2M module, where more than half (>14 genes of the set) are expressed. Additionally STMN1 + the G2M module can be used to confirm proliferating. Alternatively tumor macrophage can be biopsied and assed for expression of at least two of CD68, MERTK, and K167 and/or PCNA.

As used herein, an M1/M2 hybrid phenotype, refers to a phenotype induced in macrophages by therapeutic, such as an immunostimulatory bacterium provided herein that is attenuated by reducing or eliminating TLR2/4/5 responses to the bacterium and that encodes a non-integrating immunostimulatory payload, such as the combination of a cytokine and a STING protein that constitutively induces type I IFN. The phenotype is a proliferating and phagocytic macrophage and is associated, for example, with a combination of at least two, generally at least 3 of the following markers:

Hybrid Markers (lower than M2, higher than Ml): SPP1, CD209, CD206, such as CD209 and CD206, and/or Two or more Induced Markers: MERTK, C1QC, IFN-a2a, IFNpl, CXCL10, 4-1BBL (TNFSF9), MYC The tables in the Summary above, and examples compare macrophage phenotypes before and after administration of the exemplary therapeutic STACT encoding IL- 15/IL-15R alpha chain complex + eSTING (constitutive STING), such as the chimeric human STING with gain-of-function mutations and Tasmanian devil CTT.

As used herein, recitation of “expression of type I IFN in the macrophage is not inhibited” upon introduction of a therapeutic, delivery vehicle, or nucleic acid, means that the type I IFN occurs in the macrophage at level that is higher than in the macrophage prior to introduction of the therapeutic, delivery vehicle, or nucleic acid.

As used herein, an “expression cassette” refers to a nucleic acid construct that includes regulatory sequences for gene expression, operatively linked to nucleic acid encoding open reading frames (ORFs) that encode payloads, such as therapeutic products, or other proteins.

As used herein, 2A peptides are 18-22 amino-acid (aa)-long viral oligopeptides that mediate cleavage of polypeptides during translation in eukaryotic cells. The designation “2A” refers to a specific region of the viral genome, and different viral 2 As have generally been named after the virus they were derived from. Exemplary of these are F2A (foot-and-mouth disease virus 2A), E2A (equine rhinitis A virus), P2A (porcine teschovirus-1 2A), and T2A (Thosea asigna virus 2A). See, e.g., Liu et al. (2017) Scientific Reports 7:2193, Fig. 1, for encoding sequences. See, also, SEQ ID NOs:327-330. These peptides generally share a core sequence motif of DxExNPGP, and occur in a large number of viral families. They help break apart polyproteins by causing the ribosome to fail at making a peptide bond. The 2A peptides provide for multicistronic/polycistronic vectors, in which a plurality of proteins are expressed from a single open reading frame (ORF). For purposes herein, the 2A peptides include those that are naturally occurring, and any modified forms thereof, such as any having at least 97%, 98%, or 99% sequence identity with any naturally-occurring 2A peptide, including those disclosed herein, that result in single polypeptides being transcribed and translated from a transcript comprising a plurality (2 or more) of open reading frames.

As used a Cap Independent Translation Enhancer (CITE) sequence is a eukaryotic translation element that is part of an RNA molecule transcribed in a bacterium such that the RNA is not translated in the bacterium, but is translated in animal cells (see, g U.S. Patent No.6, 500,419). CITE sequences designed for transcription of RNA that can be translated in eukaryotic cells, but not bacterial cells, are commercially available. Exemplary is one that corresponds to the nucleotide sequence from nucleotide 2416 to nucleotide 2914 of pCITE-1 (Novagen, Inc., Madison, Wis.).

As used herein, an “interferonopathy” refers to a disorder associated with an upregulation of interferon by virtue of a mutation in a gene product involved in a pathway that regulates or induces expression of interferon. The activity of the products normally is regulated by a mediator, such as cytosolic DNA or RNA or nucleotides; when the protein product is mutated, the activity is constitutive. Type I interferonopathies include a spectrum of conditions, including the severe forms of Aicardi-Goutieres Syndrome (AGS), and the milder Familial Chilblain Lupus (FCL). Nucleic acid molecules encoding mutated products with these properties can be produced in vitro, such as by selecting for mutations that result in a gain-of-function in the product, compared to the product of an allele that has normal activity, or has further gain-of-function compared to the disease-associated gain-of-function mutants described herein.

As used herein, a “gain-of-function mutation” is one that increases the activity of a protein compared to the same protein that does not have the mutation. For example, if the protein is a receptor, it will have increased affinity for a ligand; if it is an enzyme, it will have increased activity, including constitutive activity. In particular, with respect to products, such as STING, IRF3, IRF7, MDA5, RIG-I, a constitutively active product is one that is active in the absence of a its activating ligands, such as cGAS for STING, and/or in the absence of cytosolic nucleic acids, such as DNA, RNA, nucleotides, dinucleotides, cyclic nucleotides and/or cyclic dinucleotides or other nucleic acid molecules, that lead to production of type I interferon. These nucleic acid molecules in the cytosol occur from viral or bacterial infection and/or radiation or other such exposure, leading to activation of an immune response in a host against such pathogen.

As used herein, an “origin of replication” is a sequence of DNA at which replication is initiated on a chromosome, or a plasmid, or in a virus. For small DNA, including bacterial plasmids and small viruses, a single origin is sufficient.

The origin of replication determines the vector copy number, which depends upon the selected origin of replication. For example, if the expression vector is derived from the low-copy-number plasmid pBR322, the copy number is between about 15-20 copies/cell, and if derived from the high-copy-number plasmid pUC, it can be 500-700 copies/cell. As used herein, medium copy number of a plasmid in cells is about or is 150 or less than 150, and low copy number is 5-30, such as 20 or less than 20. Low to medium copy number is less than 150 copies/cell. High copy number is greater than 150 copies/cell.

As used herein, a “CpG motif’ is a pattern of bases that includes an unmethylated central CpG (“p” refers to the phosphodiester link between consecutive C and G nucleotides), surrounded by at least one base flanking (on the 3' and the 5' side of) the central CpG. A CpG oligodeoxynucleotide is an oligodeoxynucleotide that is at least about ten nucleotides in length and includes an unmethylated CpG. At least the C of the 5' CG 3' is unmethylated.

As used herein, a “RIG-I binding sequence” refers to a 5 ’triphosphate (5 ’ppp) structure directly, or that which is synthesized by RNA polymerase III from a poly(dA-dT) sequence, which, by virtue of interaction with RIG-I, can activate type I IFN via the RIG-I pathway. The RNA includes at least four A ribonucleotides (A-A- A-A); it can contain 4, 5, 6, 7, 8, 9, 10, or more. The RIG-I binding sequence is introduced into a plasmid in the bacterium for transcription into the poly(A).

As used herein, “cytokines” are a broad and loose category of small proteins (-5-20 kDa) that are important in cell signaling. Cytokines include chemokines, interferons, interleukins, lymphokines, and tumor necrosis factors. Cytokines are cell signaling molecules that aid cell to cell communication in immune responses, and stimulate the movement of cells towards sites of inflammation, infection and trauma.

As used herein, “chemokines” refer to chemoattractant (chemotactic) cytokines that bind to chemokine receptors and include proteins isolated from natural sources as well as those made synthetically, as by recombinant means or by chemical synthesis. Exemplary chemokines include, but are not limited to, IL-8, IL-10, GCP-2, GRO-a, GRO-p, GRO-y, ENA-78, PBP, CTAP III, NAP-2, LAPF-4, MIG (CXCL9), CXCL10 (IP- 10), CXCL11, PF4, SDF-la, SDF-l , SDF-2, MCP-1, MCP-2, MCP-3, MCP-4, MCP-5, MIP-la (CCL3), MIP-ip (CCL4), MIP-ly (CCL9), MIP-2, MIP-2a, MIP-3a, MIP-3P, MIP-4, MIP-5, MDC, HCC-1, ALP, Lungkine, TIM-1, Eotaxin-1, Eotaxin-2, 1-309, SCYA17, TRAC, RANTES (CCL5), DC-CK-1, lymphotactin, and fractalkine, and others known to those of skill in the art. Chemokines are involved in the migration of immune cells to sites of inflammation, as well as in the maturation of immune cells, and in the generation of adaptive immune responses.

As used herein, an “immunostimulatory protein” is a protein that exhibits or promotes an anti-tumor immune response in the tumor microenvironment. Exemplary of such proteins are cytokines, chemokines, and co-stimulatory molecules, such as, but not limited to, IFN-ot, IFN-p, GM-CSF, IL-2, IL-7, IL-12, IL-15, IL-18, IL-21, IL- 23, IL-12p70 (IL-12p40 + IL-12p35), IL-15/IL-15R alpha chain complex (also referred to herein as IL-15/IL-15Ra, IL-15Rot-IL-15sc, IL-15 complex, and other variations, set forth herein), IL-36 gamma, IL-2 that has attenuated binding to IL-2Ra, IL-2 that is modified so that it does not bind to IL-2Ra, CXCL9, CXCL10 (IP- 10), CXCL11, CCL3, CCL4, CCL5, molecules involved in the potential recruitment and/or persistence of T-cells, CD40, CD40 ligand (CD40L), 0X40, 0X40 ligand (OX40L), 4-1BB, 4-1BB ligand (4-1BBL), 4-1BBL with a deleted cytoplasmic domain (IBBLAcyt), or with a partially deleted (truncated) cytoplasmic domain, members of the B7-CD28 family, and members of the tumor necrosis factor receptor (TNFR) superfamily.

Among the immunostimulatory proteins are truncated co-stimulatory molecules, such as, for example, 4-1BBL, CD80, CD86, CD27L, B7RP1 and OX40L, each with a full or partial cytoplasmic domain deletion, for expression on an antigen- presenting cell (APC). These truncated gene products, such as those with deletions or partial deletions of the cytoplasmic domain, are truncated such that they are capable of constitutive immunostimulatory signaling to a T-cell through co-stimulatory receptor engagement, but are unable to counter-regulatory signal to the APC, due to a deleted or truncated cytoplasmic domain.

As used herein, a “cytoplasmic domain deletion” is a deletion in all, or a portion of, the amino acid residues that comprise the cytoplasmic, or intracellular, domain of the protein, where the deletion is sufficient to effect constitutive immunostimulatory signaling to a T-cell through co-stimulatory receptor engagement, and is sufficient to inhibit counter-regulatory signaling to the APC. For example, the cytoplasmic domain of human 4-1BBL (also known as TNFSF9) comprises amino acid residues 1-28 of SEQ ID NO:342. The cytoplasmic domain of human CD80 comprises amino acid residues 264-288 of the protein; the cytoplasmic domain of human CD86 comprises amino acid residues 269-329 of the protein; the cytoplasmic domain of human CD27L (also known as CD70) comprises amino acid residues 1-17 of the protein; the cytoplasmic domain of human B7RP1 (also known as ICOSLG or ICOS ligand) comprises amino acid residues 278-302 of the protein; and the cytoplasmic domain of human OX40L (also known as TNFSF4 or CD252) comprises amino acid residues 1-23 of the protein.

As used herein, a “decoy receptor” is a receptor that can specifically bind to specific growth factors or cytokines efficiently, but is not structurally able to signal or activate the intended receptor complex. The decoy receptor acts as an inhibitor by binding to a ligand and preventing it from binding to its cognate receptor.

For example, TGF-P family receptors include the cell-surface serine/threonine kinase receptors type I (TpRI or TGFpRl) and type II (TpRII or TGFPR2), which form heteromeric complexes in the presence of dimerized ligands, as well as the type III receptor betaglycan (TpRIII or TGFPR3). Soluble decoy receptors for TGF-P, which prevent the binding of TGF-P to its receptors, include the soluble extracellular domains (the TGF-P binding regions) of TpRI, TpRII, or TpRIII (p-glycan), which can be fused with other molecules, such as an Fc domain. Additionally, BAMBI (bone morphogenetic protein (BMP) and activin membrane-bound inhibitor) is structurally related to type I receptors and acts as a decoy that inhibits receptor activation. A dominant negative TGFPR2 (dnTGFpRII), which comprises the extracellular domain of TGFPR2 and the transmembrane region, but which lacks the cytoplasmic domain required for signaling, also can be used as a TGF-p decoy receptor (see, e.g., International Application Publication No. WO 2018/138003).

As used herein, a co-stimulatory molecule agonist is a molecule that, upon binding to the co-stimulatory molecule, activates it or increases its activity. For example, the agonist can be an agonist antibody. CD40 agonist antibodies include, for example, CP-870,893, dacetuzumab, ADC-1013 (mitazalimab), and Chi Lob 7/4.

As used herein, a cytosolic DNA/RNA sensor pathway is one that is initiated by the presence of DNA, RNA, nucleotides, dinucleotides, cyclic nucleotides and/or cyclic dinucleotides or other nucleic acid molecules, that leads to production of type I interferon. The nucleic acid molecules in the cytosol occur from viral or bacterial or radiation or other such exposure, leading to activation of an immune response in a host.

As used herein, a “type I interferon pathway protein” is a protein that induces an innate immune response, such as the induction of type I interferon. As used herein, a “cytosolic DNA/RNA sensor,” is a protein that is part of a cytosolic DNA/RNA sensor pathway that leads to expression of an immune response mediator, such as type I interferon. A “cytosolic DNA/RNA sensor,” includes type I interferon pathway proteins. For example, as described herein and known to those of skill in the art, cytosolic DNA is sensed by cGAS, leading to the production of cGAMP and subsequent STING/TBK1/IRF3 signaling, and type I IFN production. Bacterial cyclic dinucleotides (CDNs, such as bacterial cyclic di-AMP) also activate STING. Hence, STING is an immunomodulatory protein that induces type I interferon. 5’ - triphosphate RNA and double stranded RNA are sensed by RIG-I and either MDA-5 alone, or MDA-5/LGP2. This leads to polymerization of mitochondrial MAVS (mitochondrial antiviral-signaling protein), and also activates TANK-binding kinase 1 (TBK1) and interferon regulatory factor 3 (IRF3). The proteins in such pathways are immunostimulatory proteins, and lead to the expression of innate immune response mediators, such as type I interferon. The immunomodulatory proteins in the DNA/RNA sensor pathways can be modified so that they have increased activity, or act constitutively in the absence of cytosolic nucleic acids and/or activating/stimulating ligands, to lead to the immune response, such as the expression of type I interferon.

As used herein, the “carboxy-terminal tail” or “C-terminal tail” (CTT) of the innate immune protein STING refers to the C-terminal portion of a STING protein that, in a wild-type STING protein, is tethered to the cGAMP -binding domain by a flexible linker region. The CTT includes an IRF3 binding site, a TBK1 binding site, and a TRAF6 binding site. STING promotes the induction of interferon beta (IFN- ) production via the phosphorylation of the STING protein C-terminal tail (CTT) by TANK-binding kinase 1 (TBK1). The interaction between STING and TBK1 is mediated by an evolutionarily conserved stretch of eight amino-acid residues in the carboxy-terminal tail (CTT) of STING. TRAF6 catalyzes the formation of K63 -linked ubiquitin chains on STING, leading to the activation of the transcription factor NF-KB and the induction of an alternative STING-dependent gene expression program. Deletion or disruption of the TRAF6 binding site in the CTT can reduce activation of NF-KB signaling. Substitution of the human STING CTT (or portions thereof), with the CTT (or corresponding portion thereof) from the STING protein of a species with low NF-KB activation, can decrease the NF-KB activation by the resulting modified human STING protein. The STING CTT is an unstructured stretch of ~40 amino acids that contains sequence motifs required for STING phosphorylation and recruitment of IRF3 (see, de Oliveira Mann et al. (2019) Cell Reports 27:1165-1175). Human STING residue S366 has been identified as a primary TBK1 phosphorylation site that is part of an LxIS motif shared among innate immune adaptor proteins that activate interferon signaling (see, de Oliveira Mann et al. (2019) Cell Reports

27 165-1175). The human STING CTT contains a second PxPLR motif that includes the residue L374, which is required for TBK1 binding; the LxIS and PxPLR sequences are conserved among vertebrate STING alleles (see, de Oliveira Mann et al. (2019) Cell Reports 27: 1165-1175). Exemplary STING CTT sequences, and the IRF3, TBK1, and TRAF6 binding sites, are set forth in the following table:

As used herein, a bacterium that is modified so that it “induces less cell death in tumor-resident immune cells” or “induces less cell death in immune cells” is one that is less toxic than the bacterium without the modification, or one that has reduced virulence compared to the bacterium without the modification. Exemplary of such modifications are those that eliminate pyroptosis in phagocytic cells and that alter lipopolysaccharide (LPS) profiles on the bacterium These modifications include disruption of or deletion of flagellin genes, pagP, or one or more components of the SPI-1 pathway, such as hilA, rod protein (e.g., prgJ), needle protein (e.g., rgl). and QseC.

As used herein, a bacterium that is “modified so that it preferentially infects tumor-resident immune cells” or “modified so that it preferentially infects immune cells” has a modification in its genome that reduces its ability to infect cells other than immune cells. Exemplary of such modifications are modifications that disrupt the type III secretion system or type IV secretion system or other genes or systems that affect the ability of a bacterium to invade a non-immune cell. For example, modifications include disrupt! on/deletion of an SPI-1 component, which is needed for infection of cells, such as epithelial cells, but does not affect infection of immune cells, such as phagocytic cells, by Salmonella.

As used herein, a “modification” is in reference to modification of a sequence of amino acids of a polypeptide, or a sequence of nucleotides in a nucleic acid molecule, and includes deletions, insertions, and replacements of amino acids or nucleotides, respectively. Methods of modifying a polypeptide are routine to those of skill in the art, such as by using recombinant DNA methodologies.

As used herein, a modification to a bacterial genome, or to a plasmid, or to a gene, includes deletions, replacements, and insertions of nucleic acid.

As used herein, RNA interference (RNAi) is a biological process in which RNA molecules inhibit gene expression or translation, by neutralizing targeted mRNA molecules to inhibit translation, and thereby expression, of a targeted gene.

As used herein, RNA molecules that act via RNAi are referred to as inhibitory by virtue of their silencing of the expression of a targeted gene. Silencing expression means that expression of the targeted gene is reduced, or suppressed, or inhibited.

As used herein, gene silencing via RNAi is said to inhibit, suppress, disrupt, or silence expression of a targeted gene. A targeted gene contains sequences of nucleotides that correspond to the sequences in the inhibitory RNA, whereby the inhibitory RNA silences expression of target mRNA.

As used herein, inhibiting, suppressing, disrupting, or silencing a targeted gene, refers to processes that alter expression, such as translation, of the targeted gene, whereby activity or expression of the product encoded by the targeted gene is reduced. Reduction, includes a complete knock-out or a partial knockout, whereby, with reference to the immunostimulatory bacteria provided herein and administration herein, treatment is effected.

As used herein, small interfering RNAs (siRNAs) are small pieces of doublestranded (ds) RNA, usually about 21 nucleotides long, with 3’ overhangs (2 nucleotides) at each end that can be used to “interfere” with the translation of proteins by binding to and promoting the degradation of messenger RNA (mRNA) at specific sequences. In doing so, siRNAs prevent the production of specific proteins based on the nucleotide sequences of their corresponding mRNAs. The process is called RNA interference (RNAi), and also is referred to as siRNA silencing, or siRNA knockdown.

As used herein, a short-hairpin RNA or small-hairpin RNA (shRNA) is an artificial RNA molecule with a tight hairpin turn that can be used to silence target gene expression via RNA interference (RNAi). Expression of shRNA in cells is typically accomplished by delivery of plasmids, or through viral or bacterial vectors.

As used herein, a tumor microenvironment (TME) is the cellular environment in which the tumor exists, including surrounding blood vessels, immune cells, fibroblasts, bone marrow-derived inflammatory cells, lymphocytes, signaling molecules and the extracellular matrix (ECM). Conditions that exist include, but are not limited to, increased vascularization, hypoxia, low pH, increased lactate concentration, increased pyruvate concentration, increased interstitial fluid pressure, and altered metabolites or metabolism, such as higher levels of adenosine, which are indicative of a tumor.

As used herein, an immune desert tumor or immune-excluded tumor is a tumor devoid of tumor infiltrating T-cells. Immune desert tumors are solid tumors where minimal effector immune cells infiltrate the tumor and there is a lack of immune response present in the tumor. Desert tumors are devoid tumor-infiltrating lymphocytes, CD8+ T-cells are absent from the tumor, including the parenchyma, stroma, and tumor periphery.

As used herein, “bactofection” refers to the bacteria-mediated transfer of genes or plasmid DNA into eukaryotic cells, such as mammalian cells.

As used herein, human type I interferons (IFNs) are a subgroup of interferon proteins that regulate the activity of the immune system. All type I IFNs bind to a specific cell surface receptor complex, such as the IFN-a receptor. Type I interferons include IFN-a and IFN-P, among others. Myeloid cells are the primary producers of IFN-a and IFN-P, which have antiviral activity that is involved mainly in innate immune responses. Two types of IFN-P are IFN-pi (IFNB1) and IFN-p3 (IFNB3).

As used herein, “Ml macrophage phenotype” and “M2 macrophage phenotype” refer to the two broad groups into which macrophage phenotypes are divided: Ml (classically activated macrophages) and M2 (alternatively activated macrophages). The role of Ml macrophages is to secrete pro-inflammatory cytokines and chemokines, and to present antigens, so that they participate in the positive immune response, and function as an immune monitor. The main pro-inflammatory cytokines they produce are IL-6, IL-12, and TNF-alpha. M2 macrophages primarily secrete arginase- 1, IL- 10, TGF-P, and other anti-inflammatory cytokines, which have the function of reducing inflammation, and contributing to tumor growth and immunosuppressive function. A macrophage with an Ml -like phenotype secretes pro- inflammatory cytokines, and does not have the immunosuppressive activity(ies) of an M2 macrophage. Conversion of an M2 macrophage into a macrophage with an Ml or Ml -like phenotype converts an M2 macrophage into one that is not immunosuppressive, but participates in an anti-tumor response. An M2 macrophage that is converted into a macrophage with an Ml or Ml -like phenotype exhibits the secretion/expression of more pro-inflammatory cytokines/chemokines and receptors, such as CD80 and CCR7, and chemokines, such as IFNy and CXCL10. Ml phenotypic markers include, but are not limited to, one or more of CD80, CD86, CD64, CD16, and CD32. The expression of nitric oxide synthase (iNOS) in Ml macrophages also can serve as a phenotypic marker. CD 163 and CD206 are major markers for the identification of M2 macrophages. Other surface markers for M2 -type cells also include CD68. A reduction or elimination of any of the M2 markers, and an increase in cytokines/chemokines that are indicative of Ml macrophages, reflect a conversion from an M2 phenotype into an Ml or Ml-like phenotype.

As used herein, an M1/M2 hybrid phenotype is anti-tumor phenotype that is achieved by treatment with therapeutics provided and described herein or that can be generated based on the disclosure herein. Post-treatment macrophage exhibit a hybrid of Ml and M2 phenotypes; macrophage with this phenotype have anti-tumor activity. In macrophage with hybrid M1/M2 phenotype, cell surface markers are upregulated relative to Ml macrophages and downregulated relative to an M2 phenotype, and upregulation of Ml inflammatory macrophage markers. Cell surface markers that are upregulated relative to M 1 macrophage (downregulated relative M2) are CD206 and retention of CD209; upregulated Ml markers include CD80/CD86 co-stimulation and upregulation of Ml lymph node-homing (LN-homing) CCR7. There is an upregulation of all classically-associated Ml inflammatory macrophage markers, expression of pattern recognition receptors (PRRs), scavenging and phagocytic markers: C14, CD206, CD209, CD68, and CD163, attributed to M2 macrophages. Hence the resulting phenotype is a hybrid of Ml and M2 macrophage phenotypes The sections and Examples below provide a more detailed description and exemplification of this phenotype.

As used herein, recitation that a nucleic acid or encoded RNA targets a gene means that it inhibits or suppresses or silences expression of the gene by any mechanism. Generally, such nucleic acid includes at least a portion complementary to the targeted gene, where the portion is sufficient to form a hybrid with the complementary portion.

As used herein, “deletion,” when referring to a nucleic acid or polypeptide sequence, refers to the deletion of one or more nucleotides or amino acids compared to a sequence, such as a target polynucleotide, or polypeptide, or a native or wild-type sequence.

As used herein, “insertion,” when referring to a nucleic acid or amino acid sequence, describes the inclusion of one or more additional nucleotides or amino acids, within a target, native, wild-type or other related sequence. Thus, a nucleic acid molecule that contains one or more insertions compared to a wild-type sequence, contains one or more additional nucleotides within the linear length of the sequence.

As used herein, “additions” to nucleic acid and amino acid sequences describe addition of nucleotides or amino acids onto either termini compared to another sequence.

As used herein, “substitution” or “replacement” refers to the replacing of one or more nucleotides or amino acids in a native, target, wild-type or other nucleic acid or polypeptide sequence with an alternative nucleotide or amino acid, without changing the length (as described in numbers of nucleotides or residues) of the molecule. Thus, one or more substitutions in a molecule does not change the number of nucleotides or amino acid residues of the molecule. Amino acid replacements

- 130 -

RECTIFIED SHEET (RULE 91 ) ISA/EP compared to a particular polypeptide can be expressed in terms of the number of the amino acid residue along the length of the polypeptide sequence.

As used herein, “at a position corresponding to,” or recitation that nucleotides or amino acid positions “correspond to” nucleotides or amino acid positions in a disclosed sequence, such as set forth in the Sequence Listing, refers to nucleotides or amino acid positions identified upon alignment with the disclosed sequence to maximize identity using a standard alignment algorithm, such as the GAP algorithm. By aligning the sequences, one skilled in the art can identify corresponding residues, for example, using conserved and identical amino acid residues as guides. In general, to identify corresponding positions, the sequences of amino acids are aligned so that the highest order match is obtained (see, e.g., Computational Molecular Biology, Lesk, A.M., ed., Oxford University Press, New York, 1988; Biocomputing: Informatics and Genome Projects, Smith, D.W., ed., Academic Press, New York, 1993; Computer Analysis of Sequence Data, Part I, Griffin, A.M., and Griffin, H.G., eds., Humana Press, New Jersey, 1994; Sequence Analysis in Molecular Biology, von Heinje, G., Academic Press, 1987; Sequence Analysis Primer, Gribskov, M. and Devereux, J., eds., M Stockton Press, New York, 1991; and Carrillo etal. (1988) SIAM J. Applied Math 48: 1073).

As used herein, alignment of a sequence refers to the use of homology to align two or more sequences of nucleotides or amino acids. Typically, two or more sequences that are related by 50% or more identity are aligned. An aligned set of sequences refers to 2 or more sequences that are aligned at corresponding positions and can include aligning sequences derived from RNAs, such as ESTs and other cDNAs, aligned with a genomic DNA sequence. Related or variant polypeptides or nucleic acid molecules can be aligned by any method known to those of skill in the art. Such methods typically maximize matches, and include methods, such as using manual alignments, and by using the numerous alignment programs available (e.g., BLASTP) and others known to those of skill in the art. By aligning the sequences of polypeptides or nucleic acids, one skilled in the art can identify analogous portions or positions, using conserved and identical amino acid residues as guides. Further, one skilled in the art also can employ conserved amino acid or nucleotide residues as guides to find corresponding amino acid or nucleotide residues between and among human and non-human sequences. Corresponding positions also can be based on structural alignments, for example by using computer simulated alignments of protein structure. In other instances, corresponding regions can be identified. One skilled in the art also can employ conserved amino acid residues as guides to find corresponding amino acid residues between and among human and non-human sequences.

As used herein, a “property” of a polypeptide, such as an antibody, refers to any property exhibited by a polypeptide, including, but not limited to, binding specificity, structural configuration or conformation, protein stability, resistance to proteolysis, conformational stability, thermal tolerance, and tolerance to pH conditions. Changes in properties can alter an “activity” of the polypeptide. For example, a change in the binding specificity of the antibody polypeptide can alter the ability to bind an antigen, and/or various binding activities, such as affinity or avidity, or in vivo activities of the polypeptide.

As used herein, an “activity” or a “functional activity” of a polypeptide, such as an antibody, refers to any activity exhibited by the polypeptide. Such activities can be empirically determined. Exemplary activities include, but are not limited to, the ability to interact with a biomolecule, for example, through antigen-binding, DNA binding, ligand binding, or dimerization, or enzymatic activity, for example, kinase activity, or proteolytic activity. For an antibody (including antibody fragments), activities include, but are not limited to, the ability to specifically bind a particular antigen, affinity of antigen-binding (e.g., high or low affinity), avidity of antigenbinding (e.g., high or low avidity), on-rate, off-rate, effector functions, such as the ability to promote antigen neutralization or clearance, virus neutralization, and in vivo activities, such as the ability to prevent infection or invasion of a pathogen, or to promote clearance, or to penetrate a particular tissue or fluid or cell in the body. Activity can be assessed in vitro or in vivo using recognized assays, such as ELISA, flow cytometry, surface plasmon resonance or equivalent assays to measure on-rate or off-rate, immunohistochemistry and immunofluorescence histology and microscopy, cell-based assays, and binding assays (e.g., panning assays).

As used herein, “bind,” “bound,” or grammatical variations thereof, refers to the participation of a molecule in any attractive interaction with another molecule, resulting in a stable association in which the two molecules are in close proximity to one another. Binding includes, but is not limited to, non-covalent bonds, covalent bonds (such as reversible and irreversible covalent bonds), and includes interactions between molecules such as, but not limited to, proteins, nucleic acids, carbohydrates, lipids, and small molecules, such as chemical compounds, including drugs. As used herein, “antibody” refers to immunoglobulins and immunoglobulin fragments, whether natural, or partially or wholly synthetically, such as recombinantly produced, including any fragment thereof containing at least a portion of the variable heavy chain and light region of the immunoglobulin molecule that is sufficient to form an antigen-binding site and, when assembled, to specifically bind an antigen. Hence, an antibody includes any protein having a binding domain that is homologous or substantially homologous to an immunoglobulin antigen-binding domain (antibody combining site). For example, an antibody refers to an antibody that contains two heavy chains (which can be denoted H and H’) and two light chains (which can be denoted L and L’), where each heavy chain can be a full-length immunoglobulin heavy chain or a portion thereof sufficient to form an antigen-binding site (e.g., heavy chains include, but are not limited to, VH chains, VH-CH1 chains and VH-CH1-CH2- CH3 chains), and each light chain can be a full-length light chain or a portion thereof sufficient to form an antigen-binding site (e.g. , light chains include, but are not limited to, VL chains and VL-CL chains). Each heavy chain (H and H’ ) pairs with one light chain (L and L’, respectively). Typically, antibodies minimally include all or at least a portion of the variable heavy (VH) chain and/or the variable light (VL) chain. The antibody also can include all or a portion of the constant region.

For purposes herein, the term antibody includes full-length antibodies and portions thereof including antibody fragments, such as anti-CTLA-4 antibody fragments. Antibody fragments, include, but are not limited to, Fab fragments, Fab’ fragments, F(ab’)2 fragments, Fv fragments, disulfide-linked Fvs (dsFvs), Fd fragments, Fd’ fragments, single-chain Fvs (scFvs), single-chain Fabs (scFabs), diabodies, anti-idiotypic (anti-Id) antibodies, or antigen-binding fragments of any of the above. Antibody also includes synthetic antibodies, recombinantly produced antibodies, multi-specific antibodies (e.g., bispecific antibodies), human antibodies, non-human antibodies, humanized antibodies, chimeric antibodies, and intrabodies. Antibodies provided herein include members of any immunoglobulin class (e.g., IgG, IgM, IgD, IgE, IgA and IgY), any subclass (e.g., IgGl, IgG2, IgG3, IgG4, IgAl, and IgA2), or sub-subclass (e.g., IgG2a and IgG2b). Antibodies for human therapy generally are human antibodies or are humanized.

As used herein, “antibody fragment(s)” refers to (i) monovalent and monospecific antibody derivatives that contain the variable heavy and/or light chains, or functional fragments of an antibody, and that lack an Fc part; and (ii) BiTE® antibodies (such as tandem scFvs), dual-affinity re-targeting antibodies (DARTs), other dimeric and multimeric antibodies, diabodies, and single-chain diabodies (scDBs) Thus, an antibody fragment includes, for example, a/an: Fab, Fab’, scFab, scFv, Fv fragment, nanobody (see, e.g., antibodies derived from Camelus bactriamu . Camelus dromedarius, or Lama paccos) (see, e.g., U.S. Pat. No. 5,759,808; and Stijlemans etal. Biol. Chem. 279:1256-1261), VHH, single-domain antibody

(dAb or sdAb), minimal recognition unit, single-chain diabody (scDb), BiTE® antibody, and DART antibody, and other antibody constructs that bind to antigens. Typically, the recited antibody fragments have a molecular weight below 60 kDa.

As used herein, “nucleic acid” refers to at least two linked nucleotides or nucleotide derivatives, including a deoxyribonucleic acid (DNA) and a ribonucleic acid (RNA), joined together, typically by phosphodiester linkages. Also included in the term “nucleic acid” are analogs of nucleic acids, such as peptide nucleic acid (PNA), phosphorothioate DNA, and other such analogs and derivatives, or combinations thereof. Nucleic acids also include DNA and RNA derivatives containing, for example, a nucleotide analog or a “backbone” bond other than a phosphodiester bond, for example, a phosphotriester bond, a phosphoramidate bond, a phosphorothioate bond, a thioester bond, or a peptide bond (peptide nucleic acid). The term also includes equivalents, derivatives, variants, and analogs of either RNA or DNA made from nucleotide analogs, and single-stranded (sense or antisense) and double-stranded nucleic acids. Deoxyribonucleotides include deoxyadenosine, deoxycytidine, deoxyguanosine, and deoxythymidine. For RNA, the uracil base is uridine.

As used herein, an isolated nucleic acid molecule is one which is separated from other nucleic acid molecules which are present in the natural source of the nucleic acid molecule. An “isolated” nucleic acid molecule, such as a cDNA molecule, can be substantially free of other cellular material, or culture medium when produced by recombinant techniques, or substantially free of chemical precursors or other chemicals when chemically synthesized. Exemplary isolated nucleic acid molecules provided herein include isolated nucleic acid molecules encoding an antibody or antigen-binding fragments provided herein.

As used herein, “operably linked” or “operatively linked,” with reference to nucleic acid sequences, regions, elements, or domains, means that the nucleic acid regions are functionally related to each other. It refers to a juxtaposition whereby the components so described are in a relationship permitting them to function in their intended manner. For instance, a promoter is operably linked to a coding sequence if the promoter effects or affects its transcription or expression. For example, a nucleic acid encoding a leader peptide can be operably linked to a nucleic acid encoding a polypeptide, whereby the nucleic acids can be transcribed and translated to express a functional fusion protein, wherein the leader peptide effects secretion of the fusion polypeptide. In some instances, the nucleic acid encoding a first polypeptide U'.g., a leader peptide) is operably linked to a nucleic acid encoding a second polypeptide, and the nucleic acids are transcribed as a single mRNA transcript, but translation of the mRNA transcript can result in one of two polypeptides being expressed. For example, an amber stop codon can be located between the nucleic acid encoding the first polypeptide and the nucleic acid encoding the second polypeptide, such that, when introduced into a partial amber suppressor cell, the resulting single mRNA transcript can be translated to produce either a fusion protein containing the first and second polypeptides, or can be translated to produce only the first polypeptide. In another example, a promoter can be operably linked to nucleic acid encoding a polypeptide, whereby the promoter regulates or mediates the transcription of the nucleic acid.

As used herein, “synthetic,” with reference to, for example, a synthetic nucleic acid molecule, or a synthetic gene, or a synthetic peptide, refers to a nucleic acid molecule, or a gene, or a polypeptide molecule that is produced by recombinant methods and/or by chemical synthesis methods.

As used herein, the residues of naturally occurring a-amino acids are the residues of those 20 a-amino acids found in nature which are incorporated into a protein by the specific recognition of the charged tRNA molecule with its cognate mRNA codon in humans.

As used herein, a “polypeptide” refers to two or more amino acids covalently joined. The terms “polypeptide” and “protein” are used interchangeably herein.

As used herein, a “peptide” refers to a polypeptide that is from 2 to about or 40 amino acids in length.

As used herein, an “amino acid” is an organic compound containing an amino group and a carboxylic acid group. A polypeptide contains two or more amino acids. For purposes herein, amino acids contained in the antibodies and immunostimulatory proteins provided herein, include the twenty naturally-occurring amino acids (see Table of Correspondence below), non-natural amino acids, and amino acid analogs (e.g., amino acids wherein the a-carbon has a side chain). As used herein, the amino acids, which occur in the various amino acid sequences of polypeptides appearing herein, are identified according to their well-known, three-letter or one-letter abbreviations (see Table of Correspondence below). The nucleotides, which occur in the various nucleic acid molecules and fragments, are designated with the standard single-letter designations used routinely in the art.

As used herein, “amino acid residue” refers to an amino acid formed upon chemical digestion (hydrolysis) of a polypeptide at its peptide linkages. The amino acid residues described herein are generally in the “L” isomeric form. Residues in the “D” isomeric form can be substituted for any L-amino acid residue, as long as the desired functional property is retained by the polypeptide. NH2 refers to the free amino group present at the amino terminus of a polypeptide. COOH refers to the free carboxy group present at the carboxyl terminus of a polypeptide. In keeping with standard polypeptide nomenclature described in J. Biol. Chem., 243:3557-59 (1968) and adopted at 37 C.F.R. §§ 1.821-1.822, abbreviations for amino acid residues are shown in the following Table:

Table of Correspondence

All sequences of amino acid residues represented herein by a formula have a left to right orientation in the conventional direction of amino-terminus to carboxylterminus. The phrase “amino acid residue” is defined to include the amino acids listed in the above Table of Correspondence, as well as modified, non-natural, and unusual amino acids. A dash at the beginning or end of an amino acid residue sequence indicates a peptide bond to a further sequence of one or more amino acid residues, or to an amino-terminal group such as NH2, or to a carboxyl-terminal group such as COOH.

In a peptide or protein, suitable conservative substitutions of amino acids are known to those of skill in the art and generally can be made without altering a biological activity of a resulting molecule. Those of skill in the art recognize that, in general, single amino acid substitutions in non-essential regions of a polypeptide do not substantially alter biological activity (see, e.g., Watson et al., Molecular Biology of the Gene, 4th Edition, 1987, The Benjamin/ Cummings Pub. Co., p. 224). Such substitutions can be made in accordance with the exemplary substitutions set forth in the following Table:

Exemplary Conservative Amino Acid Substitutions Other substitutions also are permissible and can be determined empirically or in accord with other known conservative or non-conservative substitutions.

As used herein, “naturally occurring amino acids” refer to the 20 L-amino acids that occur in polypeptides.

As used herein, the term “non-natural amino acid” refers to an organic compound that has a structure similar to a natural amino acid, but that has been modified structurally to mimic the structure and reactivity of a natural amino acid. Non-naturally occurring amino acids thus include, for example, amino acids or analogs of amino acids other than the 20 naturally occurring amino acids and include, but are not limited to, the D-stereoi somers of amino acids. Exemplary non-natural amino acids are known to those of skill in the art, and include, but are not limited to, 2-Aminoadipic acid (Aad), 3 -Aminoadipic acid (bAad), P-alanine/p-Amino-propionic acid (Bala), 2-Aminobutyric acid (Abu), 4-Aminobutyric acid/piperidinic acid (4Abu), 6-Aminocaproic acid (Acp), 2-Aminoheptanoic acid (Ahe), 2- Aminoisobutyric acid (Aib), 3-Aminoisobutyric acid (Baib), 2-Aminopimelic acid (Apm), 2,4-Diaminobutyric acid (Dbu), Desmosine (Des), 2,2'-Diaminopimelic acid (Dpm), 2,3 -Diaminopropionic acid (Dpr), N-Ethylglycine (EtGly), N-Ethylasparagine (EtAsn), Hydroxylysine (Hyl), allo-Hydroxylysine (Ahyl), 3-Hydroxyproline (3Hyp), 4-Hydroxyproline (4Hyp), Isodesmosine (Ide), allo-Isoleucine (Aile), N- Methylglycine, sarcosine (MeGly), N-Methylisoleucine (Melle), 6-N-Methyllysine (MeLys), N-Methylvaline (Me Vai), Norvaline (Nva), Norleucine (Nle), and Ornithine (Om).

As used herein, a DNA construct is a single-stranded or double-stranded, linear or circular DNA molecule that contains segments of DNA combined and juxtaposed in a manner not found in nature. DNA constructs exist as a result of human manipulation, and include clones and other copies of manipulated molecules.

As used herein, a DNA segment is a portion of a larger DNA molecule having specified attributes. For example, a DNA segment encoding a specified polypeptide is a portion of a longer DNA molecule, such as a plasmid or plasmid fragment, which, when read from the 5’ to 3’ direction, encodes the sequence of amino acids of the specified polypeptide.

As used herein, the term polynucleotide means a single- or double-stranded polymer of deoxyribonucleotides or ribonucleotide bases read from the 5’ to the 3’ end. Polynucleotides include RNA and DNA, and can be isolated from natural sources, synthesized in vitro, or prepared from a combination of natural and synthetic molecules. The length of a polynucleotide molecule is given herein in terms of nucleotides (abbreviated “nt”), or base pairs (abbreviated “bp”). The term nucleotides is used for single- and double-stranded molecules where the context permits. When the term is applied to double-stranded molecules, it is used to denote overall length and will be understood to be equivalent to the term base pairs. It will be recognized by those skilled in the art that the two strands of a double-stranded polynucleotide can differ slightly in length and that the ends thereof can be staggered; thus, all nucleotides within a double-stranded polynucleotide molecule cannot be paired. Such unpaired ends will, in general, not exceed 20 nucleotides in length.

As used herein, production by recombinant methods refers to the use of the well-known methods of molecular biology for expressing proteins encoded by cloned DNA.

As used herein, “heterologous nucleic acid” is nucleic acid that encodes products (i.e., RNA and/or proteins) that are not normally produced in vivo by the cell in which it is expressed, or nucleic acid that is in a locus in which it does not normally occur, or that mediates or encodes mediators that alter expression of endogenous nucleic acid, such as DNA, by affecting transcription, translation, or other regulatable biochemical processes. Heterologous nucleic acid, such as DNA, also is referred to as foreign nucleic acid. Any nucleic acid, such as DNA, that one of skill in the art would recognize or consider as heterologous or foreign to the cell in which it is expressed, is herein encompassed by heterologous nucleic acid; heterologous nucleic acid includes exogenously added nucleic acid that is also expressed endogenously. Heterologous nucleic acid is generally not endogenous to the cell into which it is introduced, but has been obtained from another cell, or prepared synthetically, or is introduced into a genomic locus in which it does not occur naturally, or its expression is under the control of regulatory sequences or a sequence that differs from the natural regulatory sequence or sequences.

Examples of heterologous nucleic acid herein include, but are not limited to, nucleic acid that encodes a protein in a DNA/RNA sensor pathway or a gain-of- function or constitutively active variant thereof, or an immunostimulatory protein, such as a cytokine, chemokine or co-stimulatory molecule, that confers or contributes to anti-tumor immunity in the tumor microenvironment. Other products, such as antibodies and fragments thereof, BiTEs®, decoy receptors, antagonizing polypeptides and RNAi, that confer or contribute to anti-tumor immunity in the tumor microenvironment, also are included. In the immunostimulatory bacteria, the heterologous nucleic acid generally is encoded on the introduced plasmid, but it can be introduced into the genome of the bacterium, such as a promoter that alters expression of a bacterial product. Heterologous nucleic acid, such as DNA, includes nucleic acid that can, in some manner, mediate expression of DNA that encodes a therapeutic product, or it can encode a product, such as a peptide or RNA, that in some manner mediates, directly or indirectly, expression of a therapeutic product.

As used herein, cell therapy involves the delivery of cells to a subject to treat a disease or condition. The cells, which can be allogeneic or autologous to the subject, are modified ex vivo, such as by infection of cells with immunostimulatory bacteria provided herein, so that they deliver or express products when introduced to a subject.

As used herein, genetic therapy involves the transfer of heterologous nucleic acid, such as DNA, into certain cells, such as target cells, of a mammal, particularly a human, with a disorder or condition for which such therapy is sought. The nucleic acid, such as DNA, is introduced into the selected target cells in a manner such that the heterologous nucleic acid, such as DNA, is expressed, and a therapeutic product(s) encoded thereby is (are) produced. Genetic therapy can also be used to deliver nucleic acid encoding a gene product that replaces a defective gene or supplements a gene product produced by the mammal or the cell in which it is introduced. The introduced nucleic acid can encode a therapeutic compound, such as a growth factor or inhibitor thereof, or a tumor necrosis factor or inhibitor thereof, such as a receptor thereof, that is not normally produced in the mammalian host or that is not produced in therapeutically effective amounts or at a therapeutically useful time. The heterologous nucleic acid, such as DNA, encoding the therapeutic product, can be modified prior to introduction into the cells of the afflicted host in order to enhance or otherwise alter the product or expression thereof. Genetic therapy can also involve delivery of an inhibitor or repressor or other modulator of gene expression.

As used herein, “expression” refers to the process by which polypeptides are produced by transcription and translation of polynucleotides. The level of expression of a polypeptide can be assessed using any method known in art, including, for example, methods of determining the amount of the polypeptide produced from the host cell. Such methods can include, but are not limited to, quantitation of the polypeptide in the cell lysate by ELISA, Coomassie blue staining following gel electrophoresis, Lowry protein assay, and Bradford protein assay.

As used herein, a “host cell” is a cell that is used to receive, maintain, reproduce and/or amplify a vector. A host cell also can be used to express the polypeptide encoded by the vector. The nucleic acid contained in the vector is replicated when the host cell divides, thereby amplifying the nucleic acid.

As used herein, a “vector” is a replicable nucleic acid from which one or more heterologous proteins can be expressed when the vector is transformed into an appropriate host cell. Reference to a vector includes those vectors into which a nucleic acid encoding a polypeptide or fragment thereof can be introduced, typically by restriction digest and ligation. Reference to a vector also includes those vectors that contain nucleic acid encoding a polypeptide, such as a modified anti-CTLA-4 antibody. The vector is used to introduce the nucleic acid encoding the polypeptide into the host cell for amplification of the nucleic acid, or for expression/display of the polypeptide encoded by the nucleic acid. The vectors typically remain episomal, but can be designed to effect integration of a gene or portion thereof into a chromosome of the genome. Also contemplated are vectors that are artificial chromosomes, such as yeast artificial chromosomes and mammalian artificial chromosomes. Selection and use of such vehicles are well-known to those of skill in the art. A vector also includes “virus vectors” or “viral vectors.” Viral vectors are engineered viruses that are operatively linked to exogenous genes to transfer (as vehicles or shuttles) the exogenous genes into cells.

As used herein, an “expression vector” includes vectors capable of expressing DNAthat is operatively linked with regulatory sequences, such as promoter regions, that are capable of effecting expression of such DNA fragments. Such additional segments can include promoter and terminator sequences, and optionally can include one or more origins of replication, one or more selectable markers, an enhancer, a polyadenylation signal, and the like. Expression vectors are generally derived from plasmid or viral DNA, or can contain elements of both. Thus, an expression vector refers to a recombinant DNA or RNA construct, such as a plasmid, a phage, recombinant virus or other vector that, upon introduction into an appropriate host cell, results in expression of the cloned DNA. Appropriate expression vectors are well- known to those of skill in the art and include those that are replicable in eukaryotic cells and/or prokaryotic cells and those that remain episomal or those which integrate into the host cell genome.

As used herein, “primary sequence” refers to the sequence of amino acid residues in a polypeptide, or the sequence of nucleotides in a nucleic acid molecule.

As used herein, “sequence identity” refers to the number of identical or similar amino acids or nucleotide bases in a comparison between a test and a reference polypeptide or polynucleotide. Sequence identity can be determined by sequence alignment of nucleic acid or protein sequences to identify regions of similarity or identity. For purposes herein, sequence identity is generally determined by alignment to identify identical residues. The alignment can be local or global. Matches, mismatches and gaps can be identified between compared sequences. Gaps are null amino acids or nucleotides inserted between the residues of aligned sequences so that identical or similar characters are aligned. Generally, there can be internal and terminal gaps. When using gap penalties, sequence identity can be determined with no penalty for end gaps (e.g., terminal gaps are not penalized). Alternatively, sequence identity can be determined without taking into account gaps, as the number of identical positions/length of the total aligned sequence x 100.

As used herein, a “global alignment” is an alignment that aligns two sequences from beginning to end, aligning each letter in each sequence only once. An alignment is produced, regardless of whether or not there is similarity or identity between the sequences. For example, 50% sequence identity based on “global alignment” means that in an alignment of the full sequence of two compared sequences each of 100 nucleotides in length, 50% of the residues are the same. It is understood that global alignment also can be used in determining sequence identity even when the length of the aligned sequences is not the same. The differences in the terminal ends of the sequences will be taken into account in determining sequence identity, unless the “no penalty for end gaps” is selected. Generally, a global alignment is used on sequences that share significant similarity over most of their length. Exemplary algorithms for performing global alignment include the Needleman- Wunsch algorithm (Needleman el al. (1970) J. Mol. Biol. 48:443-453). Exemplary programs for performing global alignment are publicly available and include the Global Sequence Alignment Tool available at the National Center for Biotechnology Information (NCBI) website (ncbi.nlm.nih.gov/), and the program available at deepc2.psi.iastate.edu/aat/align/align.html. As used herein, a “local alignment” is an alignment that aligns two sequences, but only aligns those portions of the sequences that share similarity or identity. Hence, a local alignment determines if sub-segments of one sequence are present in another sequence. If there is no similarity, no alignment will be returned. Local alignment algorithms include BLAST, or the Smith-Waterman algorithm (Adv. Appl. Math. 2:482 (1981)). For example, 50% sequence identity based on “local alignment” means that in an alignment of the full sequence of two compared sequences of any length, a region of similarity or identity of 100 nucleotides in length has 50% of the residues that are the same in the region of similarity or identity.

For purposes herein, sequence identity can be determined by standard alignment algorithm programs used with default gap penalties established by each supplier. Default parameters for the GAP program can include: (1) a unary comparison matrix (containing a value of 1 for identities and 0 for non-identities) and the weighted comparison matrix of Gribskov et al. (1986) Nucl. Acids Res. 14:6745- 6763, as described by Schwartz and Dayhoff, eds., Atlas of Protein Sequence and Structure, National Biomedical Research Foundation, pp. 353-358 (1979); (2) a penalty of 3.0 for each gap and an additional 0.10 penalty for each symbol in each gap; and (3) no penalty for end gaps. Whether any two nucleic acid molecules have nucleotide sequences, or any two polypeptides have amino acid sequences that are at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, or 99% “identical,” or other similar variations reciting a percent identity, can be determined using known computer algorithms based on local or global alignment (see, e.g., wikipedia.org/wiki/Sequence_alignment_software, providing links to dozens of known and publicly available alignment databases and programs). Generally, for purposes herein sequence identity is determined using computer algorithms based on global alignment, such as the Needleman-Wunsch Global Sequence Alignment tool available from NCBI/BLAST (blast.ncbi.nlm.nih.gov/Blast.cgi?CMD=Web&Page_TYPE=Blas tHome); LAlign (William Pearson implementing the Huang and Miller algorithm (Adv. Appl. Math. (1991) 12:337-357)); and the program from Xiaoqui Huang available at deepc2.psi.iastate.edu/aat/align/align.html. Typically, the full-length sequence of each of the compared polypeptides or nucleotides is aligned across the full-length of each sequence in a global alignment. Local alignment also can be used when the sequences being compared are substantially the same length. Therefore, as used herein, the term “identity” represents a comparison or alignment between a test and a reference polypeptide or polynucleotide. In one nonlimiting example, “at least 90% identical to” refers to percent identities from 90% to 100% relative to the reference polypeptide or polynucleotide. Identity at a level of 90% or more is indicative of the fact that, assuming for exemplification purposes a test and reference polypeptide or polynucleotide length of 100 amino acids or nucleotides are compared, no more than 10% (i.e., 10 out of 100) of amino acids or nucleotides in the test polypeptide or polynucleotide differ from those of the reference polypeptide or polynucleotide. Similar comparisons can be made between a test and reference polynucleotide. Such differences can be represented as point mutations randomly distributed over the entire length of an amino acid sequence, or they can be clustered in one or more locations of varying length up to the maximum allowable, e.g., 10/100, amino acid differences (approximately 90% identity). Differences also can be due to deletions or truncations of amino acid residues. Differences are defined as nucleic acid or amino acid substitutions, insertions or deletions. Depending on the length of the compared sequences, at the level of homologies or identities above about 85-90%, the result can be independent of the program and gap parameters set; such high levels of identity can be assessed readily, often without relying on software.

As used herein, a “disease or disorder” refers to a pathological condition in an organism resulting from a cause or condition, including, but not limited to, infections, acquired conditions, and genetic conditions, and that is characterized by identifiable symptoms.

As used herein, “treating” a subject with a disease or condition means that the subject’s symptoms are partially or totally alleviated, or remain static following treatment.

As used herein, “treatment” refers to any effects that ameliorate symptoms of a disease or disorder. Treatment encompasses prophylaxis, therapy and/or cure. Treatment also encompasses any pharmaceutical use of any immunostimulatory bacterium or composition provided herein.

As used herein, “prophylaxis” refers to prevention of a potential disease and/or a prevention of worsening of symptoms or of progression of a disease.

As used herein, “prevention” or prophylaxis, and grammatically equivalent forms thereof, refers to methods in which the risk or probability of developing a disease or condition is reduced and/or the severity or symptoms thereof is/are reduced.

As used herein, a “pharmaceutically effective agent” includes any therapeutic agent or bioactive agent, including, but not limited to, for example, anesthetics, vasoconstrictors, dispersing agents, and conventional therapeutic drugs, including small molecule drugs and therapeutic proteins.

As used herein, a “therapeutic effect” means an effect resulting from treatment of a subject that alters, typically improves or ameliorates, the symptoms of a disease or condition, or that cures a disease or condition.

As used herein, a “therapeutically effective amount” or a “therapeutically effective dose” refers to the quantity of an agent, compound, material, or composition containing a compound that is at least sufficient to produce a therapeutic effect following administration to a subject. Hence, it is the quantity necessary for preventing, curing, ameliorating, arresting, or partially arresting, a symptom of a disease or disorder.

As used herein, “therapeutic efficacy” refers to the ability of an agent, compound, material, or composition containing a compound to produce a therapeutic effect in a subject to whom the agent, compound, material, or composition containing a compound has been administered.

As used herein, a “prophylactically effective amount” or a “prophylactically effective dose” refers to the quantity of an agent, compound, material, or composition containing a compound, that when administered to a subject, will have the intended prophylactic effect, e.g., preventing or delaying the onset or reoccurrence, of disease or symptoms, reducing the likelihood of the onset or reoccurrence, of disease or symptoms, or reducing the incidence of viral infection. The full prophylactic effect does not necessarily occur by administration of one dose, and can occur only after administration of a series of doses. Thus, a prophylactically effective amount can be administered in one or more administrations.

As used herein, amelioration of the symptoms of a particular disease or disorder by a treatment, such as by administration of a pharmaceutical composition or other therapeutic, refers to any lessening, whether permanent or temporary, lasting or transient, of the symptoms, that can be attributed to or associated with administration of the composition or therapeutic. As used herein, an “anti-cancer agent” or “an anti -cancer therapeutic” refers to any agent or therapeutic that is destructive or toxic, either directly or indirectly, to malignant cells and tissues. For example, anti-cancer agents include agents that kill cancer cells or otherwise inhibit or impair the growth of tumors or cancer cells. Exemplary anti-cancer agents are chemotherapeutic agents, and immunotherapeutic agents.

As used herein “therapeutic activity” refers to the in vivo activity of a therapeutic product, such as a polypeptide, a nucleic acid molecule, and other therapeutic molecules. Generally, the therapeutic activity is the activity that is associated with treatment of a disease or condition.

As used herein, the term “subject” refers to an animal, including a mammal, such as a human being.

As used herein, a “patient” refers to a human subject.

As used herein, “animal” includes any animal, such as, but not limited to, primates, including humans, gorillas and monkeys; rodents, such as mice and rats; fowl, such as chickens; ruminants, such as goats, cows, deer, and sheep; and pigs and other animals. Non-human animals exclude humans as the contemplated animal. The polypeptides provided herein are from any source, animal, plant, prokaryotic and fungal. Most polypeptides are of animal origin, including mammalian origin.

As used herein, a “composition” refers to any mixture. It can be a solution, suspension, liquid, powder, paste, aqueous, non-aqueous, or any combination thereof.

As used herein, a “combination” refers to any association between or among two or more items. The combination can be two or more separate items, such as two compositions or two collections, a mixture thereof, such as a single mixture of the two or more items, or any variation thereof. The elements of a combination are generally functionally associated or related.

As used herein, “combination therapy” refers to administration of two or more different therapeutics. The different therapeutic agents can be provided and administered separately, sequentially, intermittently, or can be provided in a single composition.

As used herein, a “kit” is a packaged combination that optionally includes other elements, such as additional reagents and instructions for use of the combination or elements thereof, for a purpose including, but not limited to, activation, administration, diagnosis, and assessment of a biological activity or property. As used herein, a “unit dose form” refers to physically discrete units suitable for human and animal subjects and packaged individually, as is known in the art.

As used herein, a “single dosage formulation” refers to a formulation for direct administration.

As used herein, a “multi-dose formulation” refers to a formulation that contains multiple doses of a therapeutic agent and that can be directly administered to provide several single doses of the therapeutic agent. The doses can be administered over the course of minutes, hours, weeks, days, or months. Multi-dose formulations can allow dose adjustment, dose-pooling and/or dose-splitting. Because multi-dose formulations are used over time, they generally contain one or more preservatives to prevent microbial growth.

As used herein, an “article of manufacture” is a product that is made and sold. As used throughout this application, the term is intended to encompass any of the compositions provided herein contained in articles of packaging.

As used herein, a “fluid” refers to any composition that can flow. Fluids thus encompass compositions that are in the form of semi-solids, pastes, solutions, aqueous mixtures, gels, lotions, creams, and other such compositions.

As used herein, an isolated or purified polypeptide or protein (e.g., an isolated antibody or antigen-binding fragment thereof) or a biologically-active portion thereof (e.g., an isolated antigen-binding fragment), is substantially free of cellular material or other contaminating proteins from the cell or tissue from which the polypeptide or protein is derived, or substantially free from chemical precursors or other chemicals when chemically synthesized. Preparations can be determined to be substantially free if they appear free of readily detectable impurities as determined by standard methods of analysis, such as thin layer chromatography (TLC), gel electrophoresis and high performance liquid chromatography (HPLC), that are used by those of skill in the art to assess such purity, or are sufficiently pure such that further purification does not detectably alter the physical and chemical properties, such as enzymatic and biological activities, of the substance. Methods for purification of the compounds to produce substantially chemically pure compounds are known to those of skill in the art. A substantially chemically pure compound, however, can be a mixture of stereoisomers, in such instances, further purification might increase the specific activity of the compound. As used herein, a “cellular extract” or “lysate” refers to a preparation or fraction which is made from a lysed or disrupted cell. As used herein, “persistent viral infections” are those in which the virus is not cleared, but remains in specific cells of infected individuals. Persistent infections involve stages of silent and productive infection without rapidly killing or even producing excessive damage of the host cells. There are three types of overlapping persistent virus-host interactions that may be defined as latent, chronic, and slow infections. Diseases caused by persistent viral infections include acquired immunodeficiency syndrome (AIDS), AIDS-related complexes, chronic hepatitis, subacute sclerosing panencephalitis (chronic measles encephalitis), chronic papovavirus encephalitis (progressive multifocal leukoencephalopathy), spongiform encephalopathies (caused by prions), several herpesvirus-induced diseases, and some neoplasias. Viruses that cause these and other infections include, for example, herpesviruses, varicella-zoster virus (VZV), measles virus, human T-cell leukemia viruses (HTLVs), human immunodeficiency virus (HIV), human papovaviruses, human parvoviruses, human papillomaviruses, hepatitis viruses, adenoviruses, and parvoviruses.

As used herein, a “control” refers to a sample that is substantially identical to the test sample, except that it is not treated with a test parameter, or, if it is a plasma sample, it can be from a normal volunteer not affected with the condition of interest. A control also can be an internal control.

As used herein, the singular forms “a,” “an” and “the” include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to a polypeptide, comprising “an immunoglobulin domain” includes polypeptides with one or a plurality of immunoglobulin domains.

As used herein, the term “or” is used to mean “and/or” unless explicitly indicated to refer to alternatives only or the alternatives are mutually exclusive.

As used herein, ranges and amounts can be expressed as “about” a particular value or range. “About” also includes the exact amount. Hence, “about 5 amino acids” means “about 5 amino acids” and also “5 amino acids.”

As used herein, “optional” or “optionally” means that the subsequently described event or circumstance does or does not occur, and that the description includes instances where said event or circumstance occurs and instances where it does not. For example, an optionally variant portion means that the portion is variant or non-variant. As used herein, the abbreviations for any protective groups, amino acids and other compounds, are, unless indicated otherwise, in accord with their common usage, recognized abbreviations, or the IUPAC-IUB Commission on Biochemical Nomenclature (see, Biochem. (1972) 11(9):1726-1732). As used herein, the cancers or tumors are identified herein according to the known abbreviations, as set forth in the table below: Abbreviation Full name

For clarity of disclosure, and not by way of limitation, the detailed description is divided into the subsections that follow.

B. OVERVIEW OF IMMUNOSTIMULATORY BACTERIA FOR THERAPY

The recognition that bacteria have anti-cancer activity goes back to the 1800s, when several physicians observed the regression of tumors in patients infected with Streptococcus pyogenes. William Coley began the first study utilizing bacteria for the treatment of end-stage cancers, and developed a vaccine composed of S. pyogenes and Serratia marcescens, which was successfully used to treat a variety of cancers, including sarcomas, carcinomas, lymphomas and melanomas. Since then, a number of bacterial species, including Clostridium, Mycobacterium, Bifidobacterium, Listeria monocytogenes and Escherichia, have been studied as sources of anti-cancer vaccines (See, e.g., International PCT Application Publication Nos. WO 1999/013053 and WO 2001/025399; Bermudes etal. (2002) Curr. Opin. Drug Discov. Devel. 5:194-199; Patyar et al. (2010) Journal of Biomedical Science 17:21; and Pawlek et al. (2003) Lancet Oncol. 4:548-556).

As a therapeutic platform, bacteria have several advantages over other therapies such as oncolytic viruses. Some bacterial species can be engineered to be orally and systemically (intravenously; IV) administered, they propagate readily in vitro and in vivo, and they can be stored and transported in a lyophilized state. Bacterial chromosomes readily can be manipulated as they lack exons, and the complete genomes for numerous strains have been fully characterized (Feigner etal. (2016) mBio 7(5):e01220-16). Many types of bacteria are cheaper and easier to produce than viruses, and proper delivery of engineered bacteria can be favorable over viral delivery because they do not permanently integrate into host cell genomes, they preferentially infect myeloid cells over epithelial cells, and they can be rapidly eliminated by antibiotics if necessary, rendering them safe.

Provided herein are immunostimulatory bacteria that are modified to exploit these advantageous properties. The bacteria provided herein are modified so that they infect and accumulate in the tumor microenvironment, particularly in tumor-resident immune cells (myeloid cells), such as tumor-associated macrophages (TAMs), dendritic cells (DCs), and myeloid-derived suppressor cells (MDSCs), and also are designed to express and deliver high levels of therapeutic proteins and combinations, particularly complementary combinations, thereof. As described herein, the immunostimulatory bacteria provided herein can be used as vaccines to prevent and/or treat cancers and also as vaccines against pathogens, including bacterial, viral, parasitic, and other pathogens. These uses rely on the ability of the immunostimulatory bacteria provided herein to accumulate in tumor-resident macrophage, and also, when administered in directly, such as by intramuscular or inhalation, to accumulate in phagocytic cell, to promote durable immune responses and immunity. The properties of the bacteria that are for treatment of cancer also are advantageous for their use as vaccines. These properties derive from genome modifications that reduce TLR2 and TLR4 and 5 responses, and other modifications, such as auxotrophies that permit growth in vitro, but reduce or eliminate growth in vivo, and also auxotrophies that can reduce immunosuppressive effects of accumulated nutrients in the tumor microenvironment, such as adenosine, and those genome modifications that eliminate immunosuppressive effects of bacterial enzymes, such as asparaginase, which can inactivate T-cells.

The immunostimulatory bacteria provided herein have advantageous properties that are superior to existing bacterial therapies, and also cell therapies, oncolytic virus therapies, and prior bacterial therapies. The immunostimulatory bacteria provided herein, while they can be administered by any suitable route, are suitable for systemic, such as intravenous, administration. As shown and described herein, the immunostimulatory bacteria provided herein can target major immune pathways.

Provided are immunostimulatory bacteria that can be used or adapted as an anti-cancer therapeutics as well as an anti-cancer vaccines and pathogen vaccines, and also RNA delivery vehicles. The particular use can be selected based on the particular genome modifications and payloads. Provided are immunostimulatory bacteria that delivers a genetic payload encoding one or more therapeutic products, including, for example, a truncated co-stimulatory molecule (receptor or ligand; e.g., 4-1BBL, CD80/CD86, CD27L, B7RP1, OX40L) with a complete or partial cytoplasmic domain deletion, for expression on an antigen-presenting cell (APC), where the truncated gene product is capable of constitutive immunostimulatory signaling to a T- cell through co-stimulatory receptor engagement, and is unable to counter-regulatory signal to the APC, due to a deleted or truncated cytoplasmic domain. The immunostimulatory bacteria can encode a plurality of products including those that constitutively induce type I interferon (IFN) and those that stimulate anti-viral type of immune responses, such as IL- 15, particularly IL- 15 provided a IL-15/IL-15R alpha chain complex (IL-15 complex), and engineered STING proteins that constitutively induce type I IFN, and also can be modified to have reduced NF-KB signaling to eliminate or reduce undesirable inflammatory responses.

The immunostimulatory bacteria can encode and express one or more of IL-2, IL-7, IL-12p70 (IL-12p40 + IL-12p35), IL-12, IL-15, IL-15/IL-15Ra chain complex, IL-18, IL-21, IL-23, IL-36y, interferon-o, interferon-P, IL-2 that has attenuated binding to IL-2Ra, IL-2 that is modified so that it does not bind to IL-2Ra, CXCL9, CXCL10, CXCL11, CCL3, CCL4, CCL5, cytosolic DNA/RNA sensors or type I IFN pathway proteins, such as gain-of-function of constitutively active STING, IRF3, IRF7, MDA5, or RIG-I variants (that induce Type I IFN), inhibitors of TGF-beta, such as TGF-p inhibitory antibodies, TGF-beta polypeptide antagonists, and TGF- beta binding decoy receptors, antibodies and fragments thereof, such as those targeting immune checkpoints and other anti -cancer targets such as VEGF and IL-6, co-stimulatory receptors/molecules, such as 4-1BBL, including 4-1BBL with the cytoplasmic domain deleted or truncated or otherwise eliminated, and others. The immunostimulatory bacteria also can encode and express a truncated co-stimulatory molecule (e.g., 4-1BBL, CD80/CD86, CD27L, B7RP1, OX40L) with a complete or partial cytoplasmic domain deletion, for expression on an antigen-presenting cell (APC), where the truncated gene product is capable of constitutive immunostimulatory signaling to a T-cell through co-stimulatory receptor engagement, and is unable to counter-regulatory signal to the APC, due to a deleted or truncated cytoplasmic domain. Other encoded therapeutic products include those referred to as bispecific T-cell engagers (commercially available under the trademark BiTEs®), such as DLL3 X CD3 engagers, exemplified herein.

Combinations of such therapeutic products and agents can be expressed in a single therapeutic composition. By virtue of the modifications of the bacterial genome, the immunostimulatory bacteria exhibit tumor-specific localization and enrichment, and provide intravenous (IV) administration for activation of anti-tumor immune pathways that are otherwise toxic if systemically activated.

The immunostimulatory bacteria provided herein are genetically designed to be safe and to target tumors, the tumor microenvironment, and/or tumor-resident immune cells, and also to target phagocytic cells when administered as vaccines, such as by direct administration. The immunostimulatory bacteria provided herein include a combination of genomic modifications and other modifications, as well as encoded therapeutic products, that function in concert to provide immunostimulatory bacteria that accumulate in tumor-resident immune cells, and that persist sufficiently long enough to deliver therapeutic products, particularly combinations that induce or promote anti-cancer immune stimulation in tumors and the tumor microenvironment, without toxic side-effects, or with limited toxic side-effects. When delivered systemically, such as intravenously (IV), the immunostimulatory bacteria enrich in tumors, including in metastatic lesions; they provide efficient genetic transfer of immune payloads, specifically to tumor-resident myeloid cells, including tumor- associated macrophages (TAMs), myeloid-derived suppressor cells (MDSCs), and dendritic cells (DCs); they induce powerful, local immune responses, destroying tumors and vaccinating against future recurrence; and, when therapy is finished, they are naturally eliminated, such as by phagocytosis and destruction by the infected cells, or they can be destroyed rapidly by a course of antibiotics.

The immunostimulatory bacteria provided herein exhibit preferential accumulation in the tumor microenvironment and/or in tumor-resident immune cells due to a designed purine/ adenosine auxotrophy, and exhibit an inability to replicate inside of phagocytic cells. Immunostimulatory bacteria that avoid inactivation by serum complement allow for the delivery of a variety of immunotherapeutic agents and therapeutic products at high concentrations, directly within the tumor microenvironment, while minimizing toxicity to normal tissues, and are provided herein.

For example, as described in more detail, such as in section C.9., the immunostimulatory bacteria provided herein include one or more modifications of the genome that render them msbB'/pagP', which alters the lipid A in LPS, resulting in penta-acylation (wild-type lipid A has 6-7 fatty acid chains), reducing the TLR4 affinity; are adenosine/adenine auxotrophs, such as purl',- are asparaginase II" (ansB' which improves T-cell quality; are lacking in flagella (flagellin deficient); are deficient in genes/products that produce curli fimbriae, such as csgD', which, among other properties, removes curli fimbriae; and include other optional genomic modifications, such as insertions, deletions, disruptions, and any other modification, so that the encoded product(s) is(are) not produced in active form, as discussed in detail herein. The immunostimulatory bacteria include a plasmid that encodes one or more therapeutic products, particularly anti-cancer products, under control of a eukaryotic promoter. These same modifications and properties render them useful as vaccines and as delivery vehicles for RNA, for use as cancer therapeutics and cancer vaccines, and pathogen vaccines, particularly when delivered by direct administration, such as by inhalation, and intramuscular (IM) injection, and other direct routes for delivery of payloads to phagocytic cells.

The immunostimulatory bacteria provided herein include genome modifications, such as deletions, disruptions, and other alterations that result in inactive encoded product(s), such as changing the orientation of all or part of the gene, so that functional gene product(s) is/are not expressed. Among the immunostimulatory bacteria provided are those that are modified so that the resulting bacteria are msbB'lpurl' . In some embodiments, the bacteria are msbB' and purl', whereby the full length of at least the coding portion of the msbB and/or purl genes are/is deleted. The genome of the bacteria also can be modified so that the bacteria lack flagella. This is effected in bacteria that normally express flagella. In such bacteria, for example the fliC and fljB genes or other genes in Salmonella, or equivalent genes in other species to fliC and fljB, can be deleted or otherwise modified so that functional flagella are not expressed. The bacteria also can be modified so that they are adenosine auxotrophs, and/or are msbB'/pagP' . Also provided are immunostimulatory bacteria and pharmaceutical compositions containing them, where the bacteria do not express L-asparaginase II, whereby the bacteria are ansB'. Elimination of the encoded asparaginase activity improves or retains T-cell viability/activity. Therapeutic bacteria, such as inactivated or attenuated bacteria that are used as vaccines, can be improved by modifying the genome to eliminate asparaginase activity. Exemplary of such vaccines is the Bacillus Calmette-Guerin (BCG) vaccine and related vaccines, used to immunize against tuberculosis. The BCG vaccine is known to have variable effectiveness; eliminating the asparaginase can improve the effectiveness of such vaccine because the endogenous bacterial asparaginase inhibits or reduces T-cell activity.

The immunostimulatory bacteria provided herein, that deliver therapeutic products (such as constitutively active STING variants and other immunomodulatory proteins and products), to the tumor-resident myeloid cells promote adaptive immunity and enhance T-cell function. The immunostimulatory bacteria lead to a complete remodeling of the immunosuppressive tumor microenvironment, towards an adaptive anti-tumor phenotype, and away from a bacterial phenotype, which is characterized by the promotion of innate immunity and the suppression of adaptive immunity. As described herein, these properties and payloads also can be exploited for use of the bacteria as vaccines and RNA delivery platforms.

Immunostimulatory bacteria provided herein can exhibit significantly more, such as at least about 100,000-fold greater, tumor infiltration and enrichment, compared to unmodified bacteria, or compared to strain VNP20009. For example, the bacteria provided herein contain genome modifications whereby they infect macrophage in tumors (and other phagocytic cells when administered as vaccines) to deliver their payload, such as the combination of engineered STING (see discussion throughout regarding the various engineered STING proteins, such as the human STING with gain-of-function mutations, such as N154S/R284G, to render induction of type I IFN constitutive, and the CTT from a non-human STING that has lower NF- KB signaling activity, such as the CTT from Tasmanian devil, compared to human STING, and IL-15 in various forms, particularly IL-15/IL-15R alpha chain complex (IL- 15 complex). It is shown herein that this combination has synergistic effects, and results in high degree of T-cell, including CD4+ and CD8+ cells, infiltration of tumors. This combination is comprehensive immunotherapy by virtue of acting at numerous therapeutic intervention points. The immunostimulatory bacteria are consumed by tumor-activated macrophages (TAMs) delivering the plasmid into the host cells, which express the encoded products, such as the IL-15 and engineered STING.

The immunostimulatory bacteria provided herein are consumed by tumorresident immune cells, and deliver their contents, including the plasmid encoding the therapeutic products, which are expressed and produced in the immune cells and tumor microenvironment, to generate anti-tumor immunity. The bacteria that are asd and that do not include a complementing gene encoded on the plasmid for expression under host cell control, do not replicate in the host. Bacteria are provided that are thyA', by virtue of genome modifications, such as deletion, insertion, transposition, or modification resulting in inactive or missing gene product, requiring thymine supplementation for growth, similarly deliver payloads into phagocytic cells, but cannot replicate in such cells.

1. Bacterial Cancer Immunotherapy Many solid tumor types have evolved a profoundly immunosuppressive microenvironment that renders them highly refractory to approved checkpoint therapies, such as anti-CTLA-4, anti-PD-1 and anti-PD-Ll therapies. One mechanism by which tumors have evolved resistance to checkpoint therapies is through their lack of intratumoral T-cells and tumor antigen cross-presenting dendritic cells (DCs), described as T-cell excluded, non-inflamed, or “cold tumors” (Sharma et al. (2017) Cell 168(4):707-723). For the small number of patients whose tumors are T-cell inflamed and respond to checkpoint immunotherapies, they often experience severe autoimmune toxi cities, and many will eventually relapse and become checkpoint refractory (see, e.g., Buchbinder et al. (2015) J. Clin. Invest. 125:3377-3383; Hodi et al. (2010) 77. Engl. J. Med. 363(8):711-723; and Chen et al. (2015) J. Clin. Invest. 125:3384-3391). Tumors initiate multiple mechanisms to evade immune surveillance, reprogram anti-tumor immune cells to suppress immunity, exclude and inactivate antitumor T-cells, and develop emerged resistance to the targeted cancer therapies (see, e.g., Mahoney et al. (2Q S)Nat. Rev. Drug Discov. 14(8):561-584). Solving this problem will require immunotherapies that can properly inflame these tumors, and generate anti-tumor immunity that can provide long-lasting tumor regressions. In addition, intratumoral therapies are intractable and will be quite limiting in a metastatic disease setting. Systemically-administered therapies that properly inflame each individual metastatic lesion and overcome multiple pathways of immunosuppression are required. By virtue of their ability to specifically target tumor-resident immune cells, and to express multiple complementary genetic payloads/therapeutic products, the immunostimulatory bacteria provided herein are designed to address these issues.

2. Prior Therapies that Target the Tumor Microenvironment

A number of therapies that target the tumor microenvironment (TME) and attempt to promote anti-tumor immunity have been developed. Each has its own challenges and shortcomings, which are addressed by the immunostimulatory bacteria provided herein. a. Limitations of Autologous T-Cell Therapies

Several systemically-administered therapeutic platforms have been investigated clinically, with the goal of accessing the highly immunosuppressive tumor microenvironment and inducing the proper immune responses to inflame tumors and promote anti-tumor immunity. These platforms include chimeric antigen receptors T-cells (CAR-T cells), which are produced by harvesting T-cells from patients and re-engineering them to fuse the T-cell receptor to an antibody Ig variable extracellular domain specific for a particular tumor antigen. This confers upon the cells the antigen-recognition properties of antibodies with the cytolytic properties of activated T-cells (see, e.g., Sadelain et al. (2015) J. Clin. Invest. 125(9):3392-3400). Despite the promise and potency of this technology, such as the FDA approvals of the CD19 CAR-Ts tisagenlecleucel (such as those under the trademark Kymriah®) and axicabtagene ciloleucel (under the trademark Yescarta®), success has been limited to CD19 ⁺ hematopoietic malignancies, and at the cost of deadly immune-related adverse events (see, e.g., Jackson et al. (2016) Nat. Rev. Clin. Oncol. 13(6):370-383). Tumors can mutate rapidly to downregulate the targeted tumor antigens for solid tumors, including the antigen CD 19, thereby fostering immune escape (see, e.g., Mardiana et al. (2019) Sci. Transl. Med. 1 l(495):eaaw2293). There is not a plethora of tumorspecific target antigens. Solid tumor targets that are not expressed in healthy tissue are a major impediment to CAR-T therapy. Beyond that, CAR-T therapies suffer from other impediments to accessing solid tumor microenvironments, due to the lack of sufficient T-cell chemokine gradients, which are required for proper T-cell infiltration into tumors. In addition, once they have infiltrated tumors, they are rapidly inactivated (see, e.g., Brown et al. (2019) Nat. Rev. Immunol. 19(2):73-74). Should the safety of CAR-T cells be significantly improved and the efficacy expanded to solid tumors, the feasibility and costs associated with these labor-intensive therapies still limit their broader adoption. b. Viral Vaccine Platforms

Oncolytic viruses (OVs) have natural and engineered properties to induce tumor cell lysis, recruit T-cells to the tumor, and deliver genetic material that can be read by tumor cells to produce immunomodulatory proteins. For example, the oncolytic virus designated Talimogene laherparepvec (T-VEC), is a modified herpes simplex virus encoding anti-melanoma antigens and the cytokine GM-CSF (granulocyte-macrophage colony-stimulating factor), that is intratumorally administered. It is FDA-approved for metastatic melanoma (see, e.g., Bastin et al. (2016) Biomedicines 4(3):21). T-VEC has demonstrated clinical benefit for some melanoma patients, and with fewer immune toxicities than the immune checkpoint antibodies or the FDA-approved systemic cytokines, such as IL-2 and interferon- alpha (see, e.g., Kim etal. (2006) Cytokine Growth Factor Rev. 17(5):349-366; and Paul et l. (2015) Gene 567(2): 132-137).

Oncolytic viruses (OVs) possess a number of limitations as anti-cancer therapies. First, oncolytic viruses are rapidly inactivated by the human complement system in blood. It has proven difficult to deliver enough virus through systemic administration to have a desired therapeutic effect. Intratumoral delivery is limiting in a metastatic setting (where lesions are spread throughout the body), is intractable for most solid tumor types (e.g., lung and visceral lesions), and requires interventional, guided radiology for injection, which limits repeat dosing. Viruses can be difficult to manufacture at commercial scale and to store. Most OV-based vaccines, such as those based on paramyxovirus, reovirus and picornavirus, among others, have similar limitations (see, e.g., Chiocca etal. (2014) Cancer Immunol. Res. 2(4):295-300). Oncolytic viruses are inherently immunogenic and rapidly cleared from human blood, and T-cells that traffic into the tumor have a much higher affinity for viral antigens over weaker tumor neoantigens (see, e.g., Aleksic et al. (2012) Eur. J. Immunol. 42(12):3174-3179). Thus, in addition to the recognized technical limitations of the platform, OVs can have limited capacity to stimulate durable anti-tumor immunity (see, e.g., Kedl et cz/. (2003) Curr Opinion Immunol. 75:120-127; and Aleksic et al., (2012) Eur. J. Immunol. 42'3174-3179, which show that TCRs that bind viral antigens have higher HLA-A2 affinity that those that bind cancer related antigens). c. Bacterial Cancer Therapies

A number of bacterial species have demonstrated preferential replication within solid tumors when injected from a distal site in preclinical animal studies. These include, but are not limited to, species of Salmonella, Bifodobacterium, Clostridium, and Escherichia. The tumor-homing properties of the bacteria, combined with the host’s innate immune response to the bacterial infection, can mediate an antitumor response. This tumor tissue tropism reduces the size of tumors to varying degrees. One contributing factor to the tumor tropism of these bacterial species is the ability to replicate in anoxic and hypoxic environments. A number of these naturally tumor-tropic bacteria have been further engineered to increase the potency of the antitumor response (reviewed in Zu etal. (2014) Crit. Rev. Microbiol. 40(3):225-235; and Feigner et al. (2017) Microbial Biotechnology 10(5): 1074-1078). Despite proof- of-concept in animal studies, complement factors in human serum, that are not present in animal models, can inactivate the bacteria, limiting their use as therapies to treat cancer.

To be administered orally or systemically, the bacterial strains are attenuated so that they do not cause systemic disease and/or septic shock, but still maintain some level of infectivity for effective tumor colonization, and resistance to inactivation by complement. A number of different bacterial species, including Clostridium (see, e.g., Dang et al. (2001) Proc. Natl. Acad. Sci. U.S.A. 98(26): 15155-15160; U.S. Patent Publication Nos. 2017/0020931 and 2015/0147315; and U.S. PatentNos. 7,344,710 and 3,936,354), Mycobacterium (see, e.g., U.S. Patent Publication Nos. 2015/0224151 and 2015/0071873), Bifidobacterium (see, e.g., Dang et al. (2001); and Kimura et al. (1980) Cancer Res. 40:2061-2068), Lactobacillus (see, e.g., Dang et al. (2001)), Listeria monocytogenes (see, e.g., Le et al. (2012) Clin. Cancer Res. 18(3): 858-868; Starks et al. (2004) Ji Immunol. 173:420-427; and U.S. Patent Publication No. 2006/0051380) and Escherichia coli (see, e.g., U.S. Patent No. 9,320,787), have been studied as possible agents for anti-cancer therapy.

The immunostimulatory bacteria provided herein include genome modifications that address problems with prior bacteria developed for treating tumors. Modifications generally include changes in the genome that render a gene or gene product inactive. This can be effected by deleting a gene or a portion thereof, or disrupting a gene, or any other such change that results in an inactive product. The genome modifications improve the targeting to or accumulation of bacteria in the tumor microenvironment, and in particular, are designed so that the bacteria preferentially or only infect tumor-resident immune cells and do not infect healthy tissues, thereby decreasing toxicity and improving delivery of encoded products. The immunostimulatory bacteria also are designed to deliver therapeutic products, including combinations thereof, designed to eliminate immune suppressive effects of tumors, enhance a host’s anti -turn or response, and provide anti -tumor products. i. Listeria

Listeria monocytogenes, a live attenuated intracellular bacterium capable of inducing potent CD8 ⁺ T-cell priming to expressed tumor antigens in mouse models of cancer, has also been explored as a bacterial cancer vector (see, e.g., Le et al. (2012) Clin. Cancer Res. 18(3):858-868). In a clinical trial of the L. monocy/ogenes-based vaccine incorporating the tumor antigen mesothelin, together with an allogeneic pancreatic cancer-based GV AX vaccine in a prime-boost approach, a median survival of 6.1 months was noted in patients with advanced pancreatic cancer, versus a median survival of 3.9 months for patients treated with the GV AX vaccine alone (see, e.g., Le et al. (2015) J. Clin. Oncol. 33(12): 1325-1333) These results were not replicated in a larger phase 2b study, however, pointing to the difficulties in humans of subverting peripheral immune surveillance towards low affinity tumor neoantigens. L. monocytogenes also has shown limited immune responses to the encoded tumor antigens due to the requirement for bacteria to be lysed after phagocytosis, a prerequisite to efficient plasmid transfer, which has not been demonstrated to occur by L. monocytogenes in human macrophages. ii. Salmonella Species

Salmonella is exemplary of the immunostimulatory bacteria provided herein. Salmonella enterica serovar Typhimurium (S. typhimurium) is exemplary of a bacterial species for use for delivery of proteins and nucleic acids, such as anti-cancer therapeutics. S. typhimurium is a Gram-negative facultative anaerobe, which preferentially accumulates in hypoxic and necrotic areas due to the availability of nutrients from tissue necrosis, the leaky tumor vasculature, and their increased likelihood to survive in the immunosuppressed tumor microenvironment (see, e.g., Baban et al. (2010) Bioengineered Bugs l(6):385-394). As a facultative anaerobe, 5. typhimurium is able to grow under aerobic and anaerobic conditions, and is therefore able to colonize both small tumors that are less hypoxic, and large tumors that are more hypoxic. . typhimurium transmission through the fecal-oral route causes localized gastrointestinal infections. The bacterium can also enter the bloodstream and lymphatic system, infecting systemic tissues such as the liver, spleen and lungs. Systemic administration of wild-type S. typhimurium overstimulates TNF-a and IL-6, leading to a cytokine cascade and septic shock, which, if left untreated, can be fatal. As a result, pathogenic bacterial strains, such as S. typhimurium, must be attenuated to prevent systemic infection, without completely suppressing their ability to effectively colonize tumor tissues. Attenuation often is achieved by mutating a cellular structure that can elicit an immune response through pathogen pattern recognition, such as the bacterial outer membrane, or by limiting the bacterium’s ability to replicate in the absence of supplemental nutrients.

S. typhimurium is an intracellular pathogen that is rapidly taken up by phagocytic myeloid cells such as macrophages, or it can directly invade non- phagocytic cells, such as epithelial cells, through its Salmonella pathogenicity island 1 (SPI-l)-encoded type III secretion system (T3SS1). Once inside cells, it can replicate within a Salmonella-containing vacuole (SCV) through SPI-2 regulation, and can also escape into the cytosol of some epithelial cells (see, e.g., Agbor etal. (2011) Cell Microbiol. 13(12): 1858-1869; and Galan and Wolf-Watz (2006) Nature 444:567- 573). Genetically modified bacterial strains of S. typhimurium have been described as anti-tumor agents to elicit direct tumoricidal effects and/or to deliver tumoricidal molecules (see, e.g., Clairmont et al. (2000) J. Infect. Dis. 181: 1996-2002; Bermudes, D. et al. (2002) Curr. Opin. Drug Discov. Devel. 5:194-199; Zhao, M. et al. (2005) Proc. Natl. Acad. Set. U.S.A. 102:755-760; and Zhao, M. et al. (2006) Cancer Res. 66:7647-7652).

Various methods for attenuation of bacterial pathogens are known in the art. Auxotrophic mutations, for example, render bacteria incapable of synthesizing an essential nutrient, and deletions/mutations in genes such as aro,pur, gua, thy, nad and asd (see, e.g., U.S. Patent Publication No. 2012/0009153) are used. Nutrients produced by the biosynthesis pathways involving these genes are often unavailable in host cells, and as such, bacterial survival is challenging. For example, attenuation of Salmonella and other species can be achieved by deletion or disruption of the aroA gene, which is part of the shikimate pathway, connecting glycolysis to aromatic amino acid biosynthesis (see, e.g., Feigner etal. (2016) mBio 7(5):e01220-16). Deletion or disruption of aroA results in bacterial auxotrophy for aromatic amino acids and subsequent attenuation (see, e.g., U.S. Patent Publication Nos. 2003/0170276, 2003/0175297, 2012/0009153, and 2016/0369282; and International Application Publication Nos. WO 2015/032165 and WO 2016/025582). Similarly, other enzymes involved in the biosynthesis pathway for aromatic amino acids, including aroC and aroD have been deleted to achieve attenuation (see, e.g., U.S. Patent Publication No. 2016/0369282; and International Application Publication No. WO 2016/025582). For example, S. typhimurium strain SL7207 is an aromatic amino acid auxotroph (aroA' mutant), and strains Al and Al-R are leucine-arginine auxotrophs.

Mutations that attenuate bacteria also include, but are not limited to, mutations in genes that alter the biosynthesis of lipopolysaccharide (LPS), such as rfaL, rfaG, rfaH, rfaD, rfaP, rFb, rfa, msbB, htrB,firA,pagL,pagP, IpxR, arnT, eptA, and IpxP, mutations that introduce a suicide gene, such as sacB, nuk, hok, gef kil, or phlA,' mutations that introduce a bacterial lysis gene, such as hly and cly, mutations in genes that encode virulence factors, such as IsyA, pag, prg. iscA, virG, pic, and act, mutations in genes that modify the stress response, such as recA, htrA, htpR, hsp, and groEL, mutations in genes that disrupt the cell cycle, such as min, and mutations in genes that disrupt or inactivate regulatory functions, such as cya, crp, phoP/phoQ, and ompR (see, e.g., U.S. Patent Publication Nos. 2012/0009153, 2003/0170276, and 2007/0298012; U.S. Patent No. 6,190,657; International Application Publication No. WO 2015/032165; Feigner etal. (2016) Gut Microbes 7(2y.l71-177', Broadway et al. (2014) J Biotechnology 192: 177-178; Frahm et al. (2015) mBio 6(2):e00254-15; Kong etal. (2011) Infection and Immunity 79(12) 5027-503S and Kong et l. (2012) Proc. Natl. Acad. Sci. U.S.A. 109(47): 19414-19419). In general, attenuating mutations are gene deletions to prevent spontaneous compensatory mutations that might result in reversion to a virulent phenotype.

Another way to attenuate S. typhimurium for safety is to use the PhoP/PhoQ operon system, which is a typical bacterial two-component regulatory system, composed of a membrane-associated sensor kinase (PhoQ), and a cytoplasmic transcriptional regulator (PhoP) (see, e.g., Miller, S. I. et al. (1989) Proc. Natl. Acad. Sci. U.S.A. 86:5054-5058; and Groisman, E. A. et al. (1989) roc. Natl. Acad. Sci. U.S.A. 86:7077-7081). PhoP/PhoQ is required for virulence; its deletion results in poor survival of this bacterium in macrophages, and a marked attenuation in mice and humans (see, e.g., Miller, S. I. et al. (1989) roc. Natl. Acad. Sci. U.S.A. 86:5054- 5058; Groisman, E. A. et al. (1989) Proc. Natl. Acad. Sci. U.S.A. 86:7077-7081; Galan, J. E. and Curtiss, R. III. (1989) Microb. Pathog. 6:433-443; and Fields, P. I. et al. (1986) Proc. Natl. Acad. Sci. U.S.A. 83:5189-5193). PhoP/PhoQ deletion strains have been employed as vaccine delivery vehicles (see, e.g., Galan, J. E. and Curtiss, R. III. (1989) Microb. Pathog. 6:433-443; Fields, P. I. et al. (1986) Proc. Natl. Acad. Sci. U.S.A. 83:5189-5193; and Angelakopoulos, H. and Hohmann, E. L. (2000) Infect. Immun. 68:2135-2141). As described herein, however, it is disadvantageous for a strain to have limited survival in macrophages if the bacteria are not attempting to transfer plasmids.

These attenuated bacterial strains have been found to be safe in mice, pigs, and monkeys when administered intravenously (IV) (see, e.g., Zhao, M. etal. (2005) Proc. Natl. Acad. Sci. U.S.A. 102:755-760; Zhao, M. etal. (2006) Cancer Res. 66:7647-7652; Tjuvajev J. et al. (2001) J. Control. Release 74:313-315; and Zheng, L. et al. (2000) Oncol. Res. 12: 127-135), and certain live attenuated Salmonella strains have been shown to be well tolerated after oral administration in human clinical trials (see, e.g., Chatfield, S. N. et al. (1992) Biotechnology 10:888-892; DiPetrillo, M. D. et al. (1999) Vaccine 18:449-459; Hohmann, E. L. et al. (1996) J. Infect. Dis. 173: 1408-1414; and Sirard, J. C. et al. (1999) Immunol. Rev. 171 :5-26).

Other strains of S. typhimurium that have been attenuated for therapy are, for example, the leucine-arginine auxotroph A-l (see, e.g., Zhao etal. (2005) Proc. Natl. Acad. Sci. U.S.A. 102(3):755-760; Yu etal. (2012) Scientific Reports 2:436; U.S. Patent No. 8,822,194; and U.S. Patent Publication No. 2014/0178341) and its derivative AR-1 (see, e.g., Yu et al. (2012) Scientific Reports 2:436; Kawaguchi et al. (2017) Oncotarget 8(12):19065-19073; Zhao et al. (2006) Cancer Res. 66(15):7647- 7652; Zhao et l. (2012) Cell Cycle 11(1): 187-193; Tome et al. (2013) Anticancer Research 33:97-102; Murakami etal. (2017) Oncotarget 8(5):8035-8042; Liu etal. (2016) Oncotarget 7(16):22873-22882; and Binder et al. (2013) Cancer Immunol. Res. 1(2): 123-133); the aroA' mutant . typhimurium strain SL7207 (see, e.g., Guo et al. (2011) Gene Therapy 18:95-105; and U.S. Patent Publication Nos. 2012/0009153, 2016/0369282 and 2016/0184456), and its obligate anaerobe derivative YB1 (see, e.g., International Application Publication No. WO 2015/032165; Yu et al. (2012) Scientific Reports 2:436; and Leschner etal. (2009) PLoS ONE 4(8):e6692); the aroA~ aroll mu w S. typhimurium strain BRD509, a derivative of the SL1344 (wild-type) strain (see, e.g., Yoon et al. (2017) Eur. J. Cancer 70:48-61); the asd/cya~/crp' mutant A typhimurium strain y4550 (see, e.g., Sorenson et al. (2010) Biologies: Targets & Therapy 4:61-73) and the phoP'/phoQ' S. typhimurium strain LH430 (see, e.g., International Application Publication No. WO 2008/091375).

Attenuation, however, impacts the ability of the bacteria to accumulate in tumor-resident immune cells, the tumor microenvironment, and tumor cells. This problem is solved herein. The immunostimulatory bacteria, such as the Salmonella strains exemplified herein, are attenuated by virtue of modifications, that can include some of those described above, but also have other modifications and properties described herein that enhance the effectiveness as a cancer therapeutic.

Attenuated strains of S. typhimurium possess the innate ability to deliver DNA following phagocytosis and degradation (see, e.g., Weiss, S. (2003) Int. J. Med. Microbiol. 293(l):95-106). They have been used as vectors for gene therapy. For example, S. typhimurium strains have been used to deliver and express a variety of genes, including those that encode cytokines, angiogenesis inhibitors, toxins and prodrug-converting enzymes (see, e.g., U.S. Patent Publication No. 2007/0298012; Loeffler et al. (2008) Cancer Gene Ther. 15(12):787-794; Loeffler etal. (2007) Proc. Natl. Acad. Set. U.S.A. 104(31): 12879-12883; Loeffler etal. (2008) J. Natl. Cancer Inst. 100:1113-1116; Clairmont, C. et al. (2000) J. Infect. Dis. 181:1996-2002; Bermudes, D. et al. (2002) Curr. Opin. Drug Discov. Devel. 5: 194-199; Zhao, M. et al. (2005) Proc. Natl. Acad. Sci. U.S.A. 102:755-760; Zhao, M. et al. (2006) Cancer Res. 66:7647-7652; and Tjuvajev J. etal. (2001) Control. Release 74:313-315).

S. typhimurium has been modified to deliver the tumor-associated antigen (TAA) survivin (SVN) to antigen presenting cells (APCs) to prime adaptive immunity (see, e.g., U.S. Patent Publication No. 2014/0186401; and Xu et al. (2014) Cancer Res. 74(21):6260-6270). SVN is an inhibitor of apoptosis protein (IAP), which prolongs cell survival and provides cell cycle control, and is overexpressed in all solid tumors and poorly expressed in normal tissues. This technology uses SP 2 and its type III secretion system to deliver the TAAs into the cytosol of APCs, which then are activated to induce TAA-specific CD8 ⁺ T-cells and anti-tumor immunity (see, e.g., Xu et al. (2014) Cancer Res. 74(21):6260-6270). Similar to the /./.s7e/'L/-based TAA vaccines, this approach has shown promise in mouse models, but has not demonstrated effective tumor antigen-specific T-cell priming in humans.

In addition to the delivery of DNA that encodes proteins, S. typhimurium also has been used for the delivery of small interfering RNAs (siRNAs) and short hairpin RNAs (shRNAs) for cancer therapy. For example, attenuated S. typhimurium has been modified to express certain shRNAs, such as those that target the immunosuppressive gene indolamine dioxygenase (IDO). Silenced IDO expression in a murine melanoma model resulted in tumor cell death and significant tumor infiltration by neutrophils (see, e.g., Blache et al. (2012) Cancer Res. 72(24):6447-6456; International Application Publication No. WO 2008/091375; and U.S. Patent No. 9,453,227). Coadministration of this vector with a hyaluronidase showed positive results in the treatment of murine pancreatic ductal adenocarcinoma (see, e.g., Manuel etal. (2015) Cancer Immunol. Res. 3(9): 1096-1107; and U.S. Patent Publication No. 2016/0184456). In another study, an . typhimurium strain attenuated by aphoP/phoQ deletion, and expressing a signal transducer and activator of transcription 3 (STAT3)- specific shRNA, inhibited tumor growth and reduced the number of metastatic organs, extending the life of C57BL/6 mice (see, e.g., Zhang etal. (2007) Cancer Res. 67(12):5859-5864). In another example, S. typhimurium strain SL7207 has been used for the delivery of shRNA targeting CTNNB1, the gene that encodes p-catemn (see, e.g., Guo et al. (2011) Gene Therapy 18:95-105; and U.S. Patent Publication Nos. 2009/0123426 and 2016/0369282). The S. typhimurium strain VNP20009 has been used for the delivery of shRNA targeting STAT3 (see, e.g., Manuel et al. (2011) Cancer Res. 71(12):4183-4191 ; U.S. Patent Publication Nos. 2009/0208534, 2014/0186401 and 2016/0184456; and International Application Publication Nos. WO 2008/091375 and WO 2012/149364). siRNAs targeting the autophagy genes Atg5 and Beclinl have been delivered to tumor cells using 5. typhimurium strains Al- R and VNP20009 (see, e.g., Liu et al. (2016) Oncotarget 7(16):22873-22882).

It has been found, however, that these strains do not effectively stimulate an anti-tumor immune response, nor effectively colonize tumors for delivery of therapeutic doses of encoded products. Improvement of such strains was needed so that they more effectively stimulate an anti-tumor immune response. Immunostimulatory bacteria, and therapeutics, provided herein as well as the description and findings demonstrating the requisites for conversion macrophage phenotypes to an anti-tumor phenotype (M1/M2 hybrid phenotype) and the properties of such immunostimulatory bacteria and therapeutics, address this problem. Additional and alternative modifications of various bacteria have been described in published International PCT Application No. WO 2019/014398 and in U.S. Publication No. 2019/0017050 Al. The bacteria described in each of these publications, also described herein, can be modified as described herein to further improve their immunostimulatory and tumor-targeting properties. iii. VNP20009 (YS1646)

Exemplary of a therapeutic bacterium that can be used as a starting strain for modification(s) as described herein is the strain designated as VNP20009 (ATCC # 202165, YS1646). This virus was a clinical candidate. VNP20009 (ATCC # 202165, YS1646) was at least 50,000-fold attenuated for safety by deletion of the msbB and purl genes (see, e.g., Clairmont et al. (2000) J. Infect. Dis. 181 :1996-2002; Low et al. (2003) Methods in Molecular Medicine, Vol. 90, Suicide Gene Therapy. Methods and Reviews, pp. 47-59; and Lee c/ a/. (2000) International Journal of Toxicology 19:19- 25). Deletion or disruption to prevent expression of the msbB gene alters the composition of the lipid A domain of lipopolysaccharide, the major component of Gram-negative bacterial outer membranes (see, e.g., Low et al. (1999) Nat. Biotechnol. 17(1):37-41). This prevents lipopolysaccharide-induced septic shock, attenuating the bacterial strain and lowering systemic toxicity, while reducing the potentially harmful production of TNFa (see, e.g., Dinarello, C.A. (1997) Chest 112(6 Suppl):321S-329S; and Low et al. (1999) Nat. Biotechnol. 17(1):37-41). Deletion or disruption to prevent expression of the purl gene renders the bacteria auxotrophic for purines, which further attenuates the bacteria and enriches them in the tumor microenvironment (see, e.g., Pawelek et al. (1997) Cancer Res. 57:4537-4544; and Broadway et al. (2014) J. Biotechnology 192:177-178).

As described, inventors herein found that VNP20009 also is auxotrophic for the immunosuppressive nucleoside adenosine. Adenosine can accumulate to pathologically high levels in the tumor and contribute to an immunosuppressive tumor microenvironment (see, e.g., Peter Vaupel and Arnulf Mayer, Oxygen Transport to Tissue XXXVII, Advances in Experimental Medicine and Biology 876 chapter 22, pp. 177-183). In strains provided herein, adenosine auxotrophy is included either by virtue of the purl phenotype or other genome modification, in order to exploit the benefits of adenosine auxotrophy in the tumor microenvironment, which can accumulate adenosine, which is immunosuppressive. Immunostimulatory bacteria that are auxotrophic for adenosine can reduce or eliminate excess adenosine to thereby ameliorate its immunosuppressive effects.

When VNP20009 was administered into mice bearing syngeneic or human xenograft tumors, the bacteria accumulated preferentially within the extracellular components of tumors at ratios exceeding 300-1000 to 1, and demonstrated tumor growth inhibition, as well as prolonged survival compared to control mice (see, e.g., Clairmont et al. (2000) J. Infect. Dis. 181:1996-2002). VNP20009 demonstrated success in tumor targeting and tumor growth suppression in animal models, while eliciting very little toxicity (see, e.g., Broadway et al. (2014) J. Biotechnology 192:177-178; Loeffler et al. (2007) roc. Natl. Acad. Sci. U.S.A. 104(31): 12879- 12883; Luo et al. (2002) Oncology Research 12:501-508; and Clairmont et al. (2000) J. Infect. Dis. 181:1996-2002).

VNP20009 failed clinical trials. Results from the Phase 1 clinical trial in human metastatic melanoma revealed that, while VNP20009 was relatively safe and well tolerated, very limited anti-tumor activity was observed (see, e.g., Toso et al. (2002) J. Clin. Oncol. 20(1): 142-152). The use of VNP20009 resulted in no significant changes in metastatic disease burden, but it did demonstrate evidence of tumor colonization at the maximum tolerated dose (MTD). Higher doses, which

- 166 -

RECTIFIED SHEET (RULE 91) ISA/EP would be required to effect any anti-tumor activity, were not possible due to toxicity that correlated with high levels of pro-inflammatory cytokines.

The immunostimulatory bacteria provided and described herein provide numerous improvements and advantages that strain VNP20009 lacks. For exemplified strains, the VNP20009 strain (YS1646) can be used as the parental strain that is further modified by introduction of additional genome modifications, including those that eliminate flagella. The strain also is improved by completely deleting purl and/or msbB. Other genome modifications include elimination or inactivation of the curli fimbriae, such as by rendering the strain csgD and, optionally, rendering the strain thyA~, so that the bacteria do not replicate in vivo, and/or ansB~ to eliminate asparaginase activity, which inactivates or reduces activity of T-cells. These genome modifications individually and together improve the ability of the bacteria to accumulate in or infect phagocytic cells, and also to increase a host’s anti-viral type immune response. The immunostimulatory bacteria are modified to include various payloads that stimulate the immune system and/or reduce immunosuppression and provide therapeutic products and immunizing antigens and products.

The immunostimulatory bacteria deliver encoded genetic payloads in a tumorspecific manner, to tumor-resident myeloid cells. The immunostimulatory bacteria, by virtue of the genomic modifications, such as deletions or disruptions of genes and/or transpositions (or any mutation results in an inactive product encoded by a gene locus), and other modifications of the genome, exhibit reduced TLR2-, TLR4-, and TLR5-mediated inflammation, for example, by virtue of the elimination of the flagella, the modifications of the LPS, and the elimination of the curli fimbriae and reduced biofilm formation. As shown and described herein, elimination of TLR2- mediated, and also of TLR2/4/5- mediated activities and response promotes or enhances type I interferon production and/or reduces any reduction or inhibition of type I IFN that occurs when these receptors are activated or by virtue of TLR responses. The immunostimulatory bacteria enhance T-cell function and activities and effects, such as by virtue of the elimination of the expression of L-asparaginase II, and facilitate, provide, permit, and support plasmid maintenance. The bacteria accumulate in (or target) only, or substantially only, myeloid cells, particularly tumor-resident myeloid cells, providing highly efficient plasmid delivery after phagocytosis. The immunostimulatory bacteria provided herein colonize the tumor microenvironment, and can be administered systemically. The immunostimulatory bacteria provided herein exhibit at least 15-fold improved LD50 compared to VNP20009. Thus, a much higher dose, if needed, of the immunostimulatory bacteria provided herein can be administered without toxic effects, compared to VNP20009 (see, e.g., the table below in the section F.5. describing exemplary dosages and administration).

It is shown and described herein that immunostimulatory bacteria modified as described herein, including elimination of flagella, LPS modifications, and other modifications, preferentially accumulate in or target myeloid cells, particularly tumorresident myeloid cells. The Examples demonstrate that the immunostimulatory bacteria accumulate in such cells following systemic, such as intravenous, administration. The Examples also describe and show plasmid transfer from the immunostimulatory bacteria into tumor-resident myeloid cells, and durable protein expression following bacterial cell death, thereby delivering therapeutic products, including products that result in an anti-cancer response and phenotype. iv. Wild-Type Strains

Accumulation of VNP20009 in tumors results from a combination of factors including: the inherent invasiveness of the parental strain, ATCC 14028, its ability to replicate in hypoxic environments, and its requirement for high concentrations of purines that are present in the interstitial fluid of tumors. As described herein, it is not necessary to use an attenuated strain, such as VNP20009, as a starting bacterial strain. By virtue of the modifications described herein, the bacteria are rendered non-toxic or attenuated. The parental strain, ATCC 14028, or another wild-type strain, can be used as a starting strain, and modified as described herein. As described herein, the immunostimulatory bacteria provided herein, by virtue of genome modifications, infiltrate and colonize tumors and tumor-resident immune cells, and also tumorresident macrophages. When used as vaccines in subjects who do not have tumors, the bacteria accumulate in phagocytic cells, such as macrophages.

3. Limitations of Prior Bacterial Cancer Immunotherapies

As shown herein, and also evidenced in light of the knowledge in the art, TLRs inhibit or prevent induction of type I IFN. The immunostimulatory bacteria provided herein, include modifications the reduce or inhibit or eliminate TLR2/4/5 responses/activities, thereby overcoming the inhibition of type I IFN. The immunostimulatory bacteria provided herein, also include properties and/or payloads that enhance type I IFN expression or render type I IFN induction constitutive. In contrast, many classes of immunotherapies have significant limitations that limit their safety and efficacy, as well as complicated platforms that are not likely to be widely used. Bacteria, particularly those provided herein, have numerous advantageous properties for use as anti-cancer therapeutics, compared to, for example, oncolytic viruses. These include the ease with which they can be propagated, manufactured, stored, and eliminated from a host when treatment is completed. Viruses, however, also have advantageous properties, including the host response. The response to a bacterial infection is an innate inflammatory response, which is not advantageous for an anti-cancer therapeutic. The response to a viral infection is similar to an anti-cancer response. This is summarized in the following table.

A limitation of prior bacteria as a microbial anti-cancer platform, thus, derives from the specific immune program that is initiated upon sensing of bacteria, even intracellular bacteria, by the immune system, compared to viral-sensing pathways, which are more akin to anti-cancer pathways. The sensing programs that recognize viruses permit the generation of highly effective vaccines and durable adaptive immunity. Vaccinating against bacteria, however, has been met with limited success. For example, the FDA-approved vaccine for typhoid fever against Salmonella typhi is only 55% effective (see, e.g., Hart et al. (2016) PLoS ONE l l(l):e0145945), despite S. typhi containing a highly immunogenic Vi capsule and 0:9 antigen, which do not occur in less immunogenic bacterial strains, such as L. monocytogenes and S. typhimurium against which there are no vaccines.

Bacteria and viruses contain conserved structures known as Pathogen- Associated Molecular Patterns (PAMPs), which are sensed by host cell Pattern Recognition Receptors (PRRs). Recognition of PAMPs by PRRs triggers downstream signaling cascades that result in the induction of cytokines and chemokines, and initiation of a specific immune response (see, e.g., Iwasaki and Medzhitov (2010) Science 327(5963):291-295). The manner in which the innate immune system is engaged by PAMPs, and from what type of infectious agent, determines whether an appropriate innate or adaptive response is generated to combat the invading pathogen.

A class of PRRs, known as Toll Like Receptors (TLRs), recognize PAMPs derived from bacterial and viral origins, and are located in various compartments within the cell. TLRs recognize a variety of ligands, including lipopolysaccharide (TLR4), lipoproteins (TLR2), flagellin (TLR5), unmethylated CpG motifs in DNA (TLR9), double-stranded RNA (TLR3), and single-stranded RNA (TLR7 and TLR8) (see, e. ., Akira etal. (2001) Nat. Immunol. 2(8):675-680; and Kawai and Akira (2005) Curr. Opin. Immunol. 17(4):338-344). DNA and RNA-based viruses can be sensed either in host cytosolic compartments after phagocytosis, or directly in the cytosol. Type I interferons (IFN-a, IFN-P) are the signature cytokines induced by host recognition of single-stranded and double-stranded DNA and RNA, either of viral origin, or from the uptake of damaged host cell DNA. For example, the synthetic dsRNA analog polyinosinic:polycytidylic acid (poly(I:C)) is an agonist for endosomal TLR3; the more stable version, poly ICLC (such as that sold under the trademark Hiltonol®), of a dsRNA, has been in clinical development (see, e.g., Caskey et al. (2011) J. Exp. Med. 208(12):2357-2366). Similarly, single-stranded RNA (ssRNA) in the endosome is sensed by TLR7 and TLR8 (only in humans), and its known synthetic ligands, resiquimod and imiquimod, are FDA-approved topical cancer immunotherapies.

In the cytosol, double-stranded RNA (dsRNA) is sensed by RNA helicases such as retinoic acid-inducible gene I (RIG- 1) and melanoma differentiation- associated gene 5 (MDA-5), leading to induction of type I IFN (see, e.g., Ireton and Gale (2011) Viruses 3(6):906-919). The cytosolic sensor for dsDNA is mediated through Stimulator of Interferon Genes (STING), an ER-resident adaptor protein that is the central mediator for sensing cytosolic dsDNA from infectious pathogens or aberrant host cell damage (see, e.g., Barber (2011) Immunol. Rev. 243(l):99-108). STING signaling activates the TANK binding kinase 1 (TBK1)/ interferon regulatory factor 3 (IRF3) axis, and the NF-KB signaling axis, resulting in the induction of IFN-P and other pro-inflammatory cytokines and chemokines that strongly activate innate and adaptive immunity (see, e.g., Burdette et al. (2011) Nature 478(7370):515-518). Sensing of cytosolic dsDNA through STING requires cyclic GMP-AMP synthase (cGAS), a host cell nucleotidyl transferase that directly binds dsDNA, and in response, synthesizes a cyclic dinucleotide (CDN) second messenger, cyclic GMP- AMP (cGAMP), which binds and activates STING (see, e.g., Sun etal. (2013) Science 339(6121):786-79I; and Wu etal. (2013) Science 339(6121): 826-830).

STING also can bind to bacterially-derived CDNs, such as c-di-AMP produced from intracellular L. monocytogenes, or c-di-GMP from S. typhimurium. Cyclic GMP-AMP synthase (cGAS) produces a non-canonical CDN that can activate human STING alleles that are non-responsive to bacterially-derived canonical CDNs. Unlike the CDNs produced by bacteria, in which the two purine nucleosides are joined by a phosphate bridge with 3 ’-3’ linkages, the internucleotide phosphate bridge in the cGAMP synthesized by cGAS is joined by a non-canonical 2’-3’ linkage. These 2’-3’ molecules bind STING with 300-fold better affinity than bacterial 3’ -3’ c-di- GMP, and thus, are more potent physiological ligands of STING (see, e.g., Civril et al. (2013) Nature 498(7454)332-337; Diner et al. (2013) Cell Rep. 3(5): 1355-1361; Gao et al. (2013) Sci. Signal 6(269):pl 1 ; and Ablasser etal. (2013) Nature 503(7477):530-534). The cGAS/STING signaling pathway in humans appears to have evolved to preferentially respond to viral pathogens over bacterial pathogens.

Thus, viral-sensing PRRs and TLRs, such as STING and RIG-I, induce type I IFN, and the cytokines and chemokines that lead to effective T-cell mediated adaptive immunity. In the tumor setting, type I IFN signaling is required to induce T-cell trafficking chemokines, such as CXCL10, and also to activate DC cross-presentation of tumor antigens to prime CD8 ⁺ T-cells (see, e.g., Diamond et al. (2011) J. Exp. Med. 208(10):1989-2003; and Fuertes et al. (2011) J. Exp. Med. 208(10):2005-2016).

In contrast, host surveillance of bacteria, such as S. typhimurium, is largely mediated through TLR2, TLR4, and TLR5 (see, e.g., Arpaia et al. (2011) Cell 144(5):675-688). These TLRs signal through MyD88 (myeloid differentiation primary response protein 88) and TRIF (Toll/interleukin-1 receptor (TIR)-domain- containing adapter-inducing interferon- ) adaptor molecules to mediate induction of the NF-KB-dependent pro-inflammatory cytokines TNF-a and IL-6 (see, e.g., Pandey et al. (2015) Cold Spring Harb. Perspect. Biol. 7(l):a016246). S. typhimurium was shown to activate the NLRP3 inflammasome pathway, resulting in the cleavage of caspase-1 and the induction of the pro-inflammatory cytokines IL-ip and IL-18 that lead to pyroptotic cell death. Engagement of TLR2, TLR4 and TLR5, and inflammasome activation, induces chemokines and cytokines that lead to bacterial clearance by neutrophils and macrophages. Evidence that S. typhimurium is cleared by T-cells is limited, and antibodies that are generated against it are non-neutralizing (see, e.g, McSorley (2014) Immunol. Rev. 260(1): 168-182). Further, S. typhimurium has mechanisms to directly suppress T-cell function, impairing any potential antitumor T-cell response from being generated (see, e.g., Kullas et al. (2012) Cell Host Microbe 12(6)791-798). As a result, bacterial cancer therapies, such as S. typhimurium, lead to recruitment and clearance by neutrophils and macrophages, which are not the T-cells that are required to generate adaptive anti-tumor immunity. It is described and shown herein that these differences can explain why prior bacterial anti-cancer vaccines, even those harboring host tumor antigens, are poor T-cell priming vectors in humans. Many of the TLRs have been reported to induce type I interferon (IFN), but is found and described herein that this dogma is not necessarily correct with respect to TLR2/3/4/5/7 in primary human monocyte-derived macrophages. Experiments were performed using TLR agonists to assess the effects on type I IFN. The results showed that TLR3 and TLR4 agonists do not induce type I IFN in primary human monocytes unless pretreated with IFNa. Agonizing TLR2 does not induce type I IFN even with pretreatment with IFNa. In fact, TLR2 can inhibit induction of type I IFN. This is a heretofore unidentified problem with bacterial-based therapeutics.

These problems are among those addressed by the immunostimulatory bacteria provided herein. The immunostimulatory bacteria provided herein are engineered to have advantageous properties that were previously only provided by viral therapeutics, and also, to retain the advantageous properties of bacterial therapeutics. The immunostimulatory bacteria provided herein can be systemically administered, can localize to tumors, tumor-resident immune cells, and/or the tumor microenvironment, overcome immunosuppression, and properly activate anti-tumor immunity, while also limiting the autoimmune-related toxicities of existing systemic immunotherapies. The bacteria provided herein effectively localize to tumor-resident immune cells, and encode therapeutic anti-cancer products, and can encode a plurality of such products. For example, the bacteria provided herein can encode complementary therapeutic products. The immunostimulatory bacteria provided herein are modified to reduce or eliminate activation of TLR2, 4 and 5. As a result, they do not inhibit the induction of type I IFN. Such immunostimulatory bacteria provided herein encode payloads that induce type I IFN. This finding is generalized such that vaccines and delivery vehicles that are provided herein are designed so that TLR2 response is eliminated or reduced so that it does not prevent or inhibit type I IFN. In some embodiments, TLR4 and 5 responses are reduced or eliminated, thereby providing vaccines and other immune-stimulating therapeutics that do not inhibit type I IFN.

Provided herein is a superior microbial anti-cancer, vaccine, RNA delivery platform, engineered to retain the beneficial properties of bacteria, while eliciting a viral-like immune response that induces effective adaptive immunity. As described herein, bacteria, such as strains of Salmonella and other species, can be modified as described herein to have reduced inflammatory effects, and thus, to be less toxic. As a result, for example, higher dosages can be administered. Any of these strains of Salmonella, as well as other species of bacteria, known to those of skill in the art and/or listed above and herein, can be modified as described herein.

The immunostimulatory bacteria provided herein are modified to have increased colonization of the tumor microenvironment, tumor-resident immune cells, and tumors. They are engineered so that they have reduced toxicity, and other properties that target them to the tumor microenvironment, including adenosine auxotrophy. The strains provided herein also are engineered so that they are not inactivated by complement. These characteristics, particularly those that result in colonization/infection of phagocytic cells also render the bacteria useful as vaccine platforms and as RNA delivery vehicles. They can be adapted for each use by virtue of the payloads selected, the regulatory sequences employed, depending upon whether bacterial or host cell transcriptional/translational machinery is to be used for expression of encoded products, and the locus of the host cells or tissues in which the products are produced.

The bacterial strains provided herein are engineered to deliver therapeutic products. The bacterial strains herein deliver immunostimulatory proteins, including cytokines, chemokines and co-stimulatory molecules, as well as modified gain-of- function cytosolic DNA/RNA sensors that can constitutively evoke or induce type I IFN expression, and other therapeutic products, such as, but not limited to, antibodies and fragments thereof, TGF-P and IL-6 binding decoy receptors, TGF-P polypeptide antagonists, bispecific T-cell engagers (BiTEs®), RNAi, and complementary combinations thereof, that promote an anti-tumor immune response in the tumor microenvironment. The bacterial strains also include genomic modifications that reduce pyroptosis of phagocytic cells, thereby providing for a more robust immune response, and/or reduce or eliminate the ability to infect/invade epithelial cells, but retain the ability to infect/invade phagocytic cells, so that they accumulate more effectively in tumors, the tumor microenvironment and in tumor-resident immune cells. The bacterial strains also can be modified to be resistant to inactivation by complement factors in human serum. The bacterial strains also can be modified to encode therapeutic products, including, alone or in combinations, for example, cytokines, chemokines, co-stimulatory molecules, constitutively active inducers of type I IFN, and monoclonal antibodies (and fragments thereof) to immune checkpoints, and also to other such targets.

Provided is an anti-cancer therapeutic product that delivers a genetic payload encoding a truncated co-stimulatory molecule (receptor or ligand; e.g., 4-1BBL, CD80, CD86, CD27L, B7RP1, OX40L), with a full or partial cytoplasmic domain deletion, for expression on an antigen-presenting cell (APC), where the truncated gene product is capable of constitutive immuno-stimulatory signaling to a T-cell through co-stimulatory receptor engagement, and is unable to counter-regulatory signal to the APC due to a deleted or truncated cytoplasmic domain.

The bacteria also can encode antigens, such as tumor antigens, and pathogen antigens or proteins, in additional to (or in lieu of) the immunostimulatory protein(s), such as STING and cytokines.

4. Therapeutics That Induce a Hybrid M1/M2 Anti-tumor Phenotype in Tumor-Resident Macrophages

Therapeutics described herein, such as immunostimulatory bacteria provided herein, upon administration, result in macrophages that have an anti-cancer phenotype that is a hybrid M1/M2 phenotype, described below, and also depicted in the Figures. Therapeutics provided and described herein produce a new anti-tumor macrophage phenotype. This phenotype results from infection of tumor-resident macrophage with anti-cancer therapeutics that deliver a payload comprising a combination of immunostimulatory proteins, such as, a cytokine, such as an IL-15 and a modified STING (referred to as eSTING herein, exemplary modified are detailed herein, see above and below for detailed descriptions), that constitutively induces type I IFN, and optionally has lower NF -KB signaling activity compared to unmodified human STING (the sequences of allelic STING proteins of human STING set forth in SEQ ID NOs: 305-309). The encoding nucleic acid is delivered in a therapeutic that has reduced or eliminated TLR2 or TLR2/4/5 inducing activity or response, whereby the expression of type I IFN is not inhibited, and is constitutive. The encoded cytokine is one that induces an anti-viral or anti-tumor response. The therapeutic can infect proliferating macrophage, generally, as shown herein M2 macrophages, which can express the encoded payload. The nucleic acid encoded by the delivery vehicle generally is nonintegrating, such as in a non-integrating plasmid. Exemplary of the therapeutics are immunostimulatory bacteria provided herein, such as the Salmonella tryphimuium strain, designated as STACT, that encodes a payload comprising a cytokine and type I IFN- inducing protein, such as the combination IL-15/IL-15R alpha chain complex + eSTING. The STACT additionally can encode a tumor antigen, and/or other payloads, including any described herein, suitable for a particular treatment. Thus, provided are therapeutics, such a nanoparticles, viruses, exosomes, bacteria, and other delivery vehicles, that, upon administration, such as systemic or intratumoral administration, can be taken up by tumor-resident macrophages, and that encode a combination payload, whereby the resulting macrophages have a hybrid M1 M2 phenotype, as detailed and exemplified herein.

Also provided are methods for screening for such therapeutics by assaying in macrophages and detecting the M1/M2 phenotype. Also provided are combination therapies. It also is shown herein, that pre-treatment of a subject with cancer with an apoptosis-inducing chemotherapy, prior to administration of any of the immunostimulatory bacteria provided herein, leads to an improved therapeutic response to the immunostimulatory bacteria. The following discussion describes the results herein, including the requisite hybrid M1/M2 macrophage phenotype that are exemplified herein. As detailed herein and exemplified in the examples, therapeutics can produce macrophage with an M1/M2 hybrid phenotype. The resulting macrophage can phagocytose apoptotic tumor cells, and can express the encoded immunostimulatory proteins that produce an anti-cancer response in the subject.

Macrophages are the most abundant myeloid cell across solid tumor types, including, but not limited to, lymphoma, nasopharyngeal, esophageal, thyroid, lung, breast, hepatocellular carcinoma, stomach, pancreatic, kidney, colorectal, ovarian and fallopian tube carcinoma, and myeloma. Macrophage limit T-cell infiltration into solid tumors and suppress their function, such as in triple negative breast cancer; these macrophages dominate the intratumoral immune population and promote T-cell exclusion in colorectal cancer. Macrophage are the predominant producers of type I

- 175 -

RECTIFIED SHEET (RULE 91) ISA/EP IFN. It is shown herein, that immunostimulatory bacteria, provided herein, such as the exemplary strains designated STACT. STACT strains only target phagocytic tumorresident myeloid cells after tumor-specific enrichment. STACT tumor specific infiltration and enrichment, is followed by consumption by tumor-associated macrophage and ectopic gene expression.

As known to those of skill in the art Ml tumor-associated macrophages have a variety of (TAMs). Conventional wisdom holds that Ml macrophages have antitumor activity. It is shown herein that this is not correct. The M2 macrophage, not Ml, are phagocytic of apoptotic tumor cells. Ml macrophage in COVID and cancer are inflammatory and immunosuppressive because of the pro-inflammatory cytokines that are produced. Ml cytokines impair type I IFN and CD8+ T-cell priming in COVID and cancer. Therefore, conversion of M2 macrophage to Ml in tumors not desirable. It is shown herein that a phenotype that is a hybrid of Ml and M2 is a phenotype that should be produced for an anti-tumor response. Therapeutics described and provided herein result in the heretofore unknown, but advantageous, M1/M2 hybrid phenotype. Below is a table, also described in the Examples, that describes markers attributed to Ml and M2 phenotypes.

Table: Markers Attributed to Ml and M2 Macrophage Phenotypes

As shown in the Examples, bactofection with an immunostimulatory bacterium provided herein that is genome modified so that it does not induce TLR2/4/5 (or not induce sufficiently to block type I IFN production), encoding a payload of immunostimulatory proteins, alters the phenotype of the infected macrophage to a phenotype that herein is referred to as an M1/M2 hybrid phenotype. The following table describes markers associated with this phenotype. Table: Markers Attributed to Ml and M2 Macrophage Phenotypes Post- Treatment

It is shown herein that treatment, such as with immunostimulatory bacteria provided herein that are designed so that they infect tumor-resident macrophages, and to encode and express immunostimulatory proteins, such as a combination of a cytokine and a STING protein that is engineered to have constitutive activity, alter the phenotype of the infected macrophage to one that is such an M1/M2 hybrid. Macrophage with this phenotype have anti-tumor activity. Cell surface markers, that are upregulated relative to Ml macrophage (downregulated relative M2) are CD206 and retention of CD209; upregulated markers include CD80/CD86 co-stimulation and upregulation of Ml lymph node-homing (LN-homing) CCR7. There is an upregulation of all classically-associated Ml inflammatory macrophage markers, expression of pattern recognition receptors (PRRs), scavenging and phagocytic markers: C14, CD206, CD209, Cd68, and CD163, attributed to M2 macrophages. Hence the resulting phenotype is a hybrid of Ml and M2 macrophage phenotypes.

It also is shown herein that the immunostimulatory bacteria that infect the tumor-resident macrophage infect M2 macrophage, not Ml, and that the encoded payload of immunostimulatory proteins, which expression results in the hybrid phenotype, only can be expressed if the macrophage are proliferating It is shown herein that it is necessary for the macrophage be proliferating for the encoded nucleic acid to get into the nucleus of the cell for transcription. The nucleic acid should be non-integrating. Thus, the process, as shown in the Examples, following administration results in infection of proliferating M2 macrophage, which then exhibit the hybrid M1/M2 phenotype, which as shown herein is an anti-cancer phenotype. Any therapeutic that can infect proliferating M2 macrophage, and deliver immunostimulatory proteins, such as IL-15/IL-15R alpha chain complex and eSTING (particularly constitutive STING, such as any described herein), and other similar combinations of immunostimulatory proteins, where they are expressed by the macrophage, result in this advantageous, heretofore unreported, phenotype. As shown and described herein, administration of a therapeutic, such as the strain designated STACT with a payload immunostimulatory protein, including one that induces type I IFN, provides for macrophage phagocytosis of apoptotic tumor cells. Results of studies show that exemplary strain STACT, IL-15/IL-15R alpha chain complex - eSTING, maintains M2-like properties of apoptotic tumor cell phagocytosis. Unlike small molecule STING agonists that polarize M2 to an Ml phenotype, the M2 macrophage treated with therapeutics, such as these immunostimulatory bacteria provided herein, retain phagocytic functions.

Type I IFN, which is encoded by the therapeutics provided herein, such as the STACT that encodes the modified STING, enhances macrophage phagocytosis through induction if interferon stimulated gene 15 (ISG15). The results show that STACT IL-15/IL-15R alpha chain complex +eSTING induction of type I IFN enhances phagocytosis of apoptotic tumor cells. Results, detailed in the Examples, show that human M2 are more phagocytic than Ml for bacteria, and bacterial phagocytosis by the macrophage is enhanced for the immunostimulatory bacteria that encode the IL-15/IL-15R alpha chain complex + eSTING. Phagocytosis of these bacteria by M2 macrophages induces the hybrid M1/M2 phenotype. Thus, it also can prime apoptotic tumor antigens (and also encode the antigens).

It is shown herein that transgene expression only occurs in proliferating cells. Factors associated with M2 polarization induce macrophage proliferation, while Ml- associated factors do not. Ml-polarizing signals and adenine and hypoxia can prevent cell cycle entry, which can results in impaired plasmid entry. The immunostimulatory bacteria provided herein are adenosine auxotrophs; they can deplete adenosine in the tumor microenvironment. It is shown in animal models, and described herein, that while there are detectable CFUs of the bacteria, there is no expression of plasmid payloads in spleen and liver. In contrast as shown in the Examples, M2 macrophages, not liver Kupffer or human vascular endothelial (HUVECs) cells, demonstrate payload delivery. It shown herein, that M2 macrophages, not Kupffer or HUVECs are phagocytic and proliferating, providing for payload delivery, which when the proper combination of immunostimulatory proteins, such as a cytokine, such a IL- 15, such as IL-15/IL-15R alpha chain complex, and a protein that constitutively induces type I IFN, such as the eSTING proteins provided herein is encoded and expressed, it leads to the desirable and advantageous anti-tumor M1/M2 hybrid phenotype. Macrophage proliferation can be assessed to monitor treatment and/or to select subjects for treatment. It is shown herein that STMN1 (stathmin 1 - microtubule destabilizer) is highly correlated with G2/M proliferation gene signature. It can be used as marker for proliferation M2 macrophages. Factors associated with M2 polarization induce macrophage proliferation, while Ml -associated factors do not. Therapeutics, such as the immunostimulatory bacteria provided herein that encode the combination of immunostimulatory proteins prime and activate tumor-associate antigen-specific CD8+ cells, and induce anti-tumor immunity.

Figure 16 depicts the priming and activation of tumor-associate antigen (TAA)- specific CD8 ⁺ T-cells and induction of anti-tumor immunity, such as by the immunostimulatory bacterium, designated as STACT, that encode immunostimulatory proteins, such as a cytokine and a protein that constitutively induces type I IFN, such as an eSTING. The resulting macrophages can phagocytose apoptotic tumor cells, and recruit and present tumor neo-antigens to CD8+ cells under co-stimulation and cytokine activation. The next figure depicts that M2/M2 antigens and properties of the resulting macrophage. As shown in the figures and demonstrated in the Examples, the resulting macrophage have Ml markers properties, such as CD80/86, CCR7, ZFNy, IFNp, CXCLI0/11, MHC I and n, and antigen Presenting, lymph-node (LN)-homing, and M2 markers and properties, such as CD 14, CD 163, CD206, and CD209, phagocytic of apoptotic tumor cells and also of bacteria, including the bacterial therapeutic. Hence, these therapeutics result in macrophage with a hybrid phenotype that provides anti-tumor immunity.

SPP1 v C1QC tumor-associated macrophages (TAMs)

Clq Complement Factor has phagocytic wound healing and anti-tumor functions. Clq belongs to the classical complement pathway bridging innate and adaptive immunity. Clq promotes macrophage phagocytosis of apoptotic cells through induction of MERTK phagocytic receptor. MERTK encodes the phagocytosis receptor required for macrophage phagocytosis of apoptotic C1QC+ TAMs in CRC have high CD80 co-stim and CXCL10 expression, and engage T-cells. CIQC ^I1 'SPP I ^I °" TAMs are associated with the best overall survival (OS) in CRC. C1QC (complement protein) promotes macrophage phagocytosis of apoptotic cells through induction of MERTK, and also is a marker of T-cell priming and migratory macrophages. SPP1 encodes Osteopontin, which promotes macrophage attachment and opsonization though the a _x p2 integrin receptors. SPP1+ macrophage are the most wound-healing macrophages. Thus there are two distinct tumor- associated macrophage (TAM) subsets: complement-activating C1QC’ TAMs and wound-healing SPPl’ TAMs. Macrophages have large variations in functional markers across a range of solid tumor types. C1QC and SPP1 subsets predominate in many tumor types. Single-cell RNA-seq analysis performed on immune and stromal populations in human and murine colorectal cancer revealed the following: two distinct TAM subsets: complement-activating C1QC TAMs, and wound-healing SPPl' TAMs. C l QC ^lo "SPP I ¹¹¹ TAMs are associated with lowest overall survival (OS) and most resistance to immunotherapy CSF1R antagonists do not deplete SPP1+ macrophages, only C1QC+, and explaining why therapy failed in the clinic.

Osteopontin (OPN) recruits monocytes and opsonizes bacteria to enhance bacterial phagocytosis; it mediates processes for cancer progression. OPN stimulates monocyte recruitment; it binds to monocytes and bacteria and enhances phagocytosis via opsonization. Monocyte recruitment and opsonization occur through the a _x 2 integrin receptor. SPP1 ⁺ TAMS are highly phagocytic and promote a wound-healing tumor microenvironment. SPPl is broadly upregulated in tumor tissue compared to healthy tissue, particularly lung, breast, head and neck, gastric and colon cancers. For example the highest myeloid/T-cell tumor subtypes in CRC (colorectal cancer) have the highest SPP1+ macrophages and poorest survival. The highest ratio SPP1/CD68 macrophages have the poorest survival. SPPl associated with poor survival across many tumor types.

C1QC is not broadly upregulated in tumor vs. healthy tissue, but is upregulated in kidney cancers (KIRC and KICH) where SPPl is not upregulated. Other cancers, such as lung, liver, and color have much loser C1QC in tumor tissue than in healthy tissue. C1QC is much less associated with poor survival than is SPPl. Primary human MDMs (monocyte derived macrophages) are C1QC+. Treatment therapeutics described herein, such as he immunostimulatory bacteria provided herein, such as those that encode a cytokine +a protein that induces type I IFN, such as IL-15/IL-15R alpha chain complex + eSTING, induces ahybrid SPP1+/ C1QC+ phenotype. As shown in the Examples, these immunostimulatory bacteria induce a phenotype in which SPP1, C1QC, and MERTK are markers. Their expression (relative to reference protein actin) is correlated with tumor Status. A higher ratio of C1QC to SPP1 is correlated with payload delivery and CD8+ T-cell infiltration For example, STACT is very efficacious in breast and colon tumor models, for example, that are C l QC ^I| 7SPP I ^lu " . Experiments described in the Examples, show that suppression of SPP1 and induction of C1QC ae correlates of strain potency. For example, proliferative macrophages in breast cancer are MK167+ C1QC+ STMN1+ (stathmin 1); proliferating macrophage in human CRC are C1QC+ STMN1+. Other experiments show that proliferating macrophages across solid tumor types are identifiable solely by their SPP1 C1QC status.

Proliferating macrophages and apoptosis

It is shown herein that nucleic acid in therapeutics that infect or enter into tumor-resident macrophages are not produced unless the macrophage are proliferating. Hence, provided herein are methods for identifying tumors that comprise proliferating macrophages for treatment with the therapeutics provided herein that result in the M1/M2 phenotype.

The Examples show that proliferating macrophage in solid tumors can be identified by their G2M cell cycle pathway score (generally > 14 indicates proliferation) and by STMN1 ⁺ . Experiments and results in the Examples also show that macrophages increase with chemotherapy. For example, in breast cancers CD68 was significantly higher after two rounds of chemotherapy. Proliferating macrophages are highest in lung colon and breast cancer. It is shown herein that gene delivery therapy requires proliferating macrophage for payload expression. A regiment for treatment with any of the immunostimulatory bacteria provided herein, can be preceded by chemotherapy.

The amount of proliferating macrophages in tumors prior to dosing with the therapeutics herein can be achieved with pre-treatment with anti-PD-1. The role that PD-1 and PD-L1 play on myeloid cell biology has been underappreciated relative to their roles in T-cell - tumor cell interactions.

The Examples also show that combination therapy regiments can be advantageous; the data show that prior treatment with chemotherapy can enhance the therapeutic effect of immunostimulatory bacteria provided here by increasing the number of proliferating macrophages in the tumors. The ability to enhance plasmid payload delivery in tumors also can be achieved through induction of apoptosis, which recruits phagocytic and proliferating macrophages that are required for DNA transfer to turn or- resident macrophages. Several types of chemotherapy have been described to promote tumor apoptosis, including docetaxel (DTX), paclitaxel (PTX), doxorubicin (DOX), 5 -fluorouracil (5-FU), carboplatin (CARB), and cyclophosphamide (CTX) (Anfray etal., (2019) Cells 9(1):46).

Immunostimulatory bacteria provided herein include adenosine auxotrophs, such as the purT strains, including the strains designated STACT. These bacteria can replicate in solid tumors because there are high levels of purine and/or adenosine. The tumor microenvironment is immunologically dysregulated permitting enrichment of immunosuppressive phagocytic cells (dendritic cells (DCs), tumor-associated macrophages (TAMs), and neutrophils/MDSCs. The tumor core generates purines and purine derivatives. Tumor cell apoptosis generates phagocyte-recruiting ATP; high metabolic turnover generates extracellular purines. Hypoxia leads to CD39/CD73 +TNAP (tissue-nonspecific alkaline phosphatase) generates extracellular adenosine. High adenosine inhibits phagocytosis by PBMC-derived macrophage phagocytosis. The presence of tumor-specific high concentrations of adenosine (generally > 10 pM) can inhibit monocyte-derived tumor-associated macrophage (TAM) recruitment and phagocytosis of the immunostimulatory bacteria and apoptotic tumor cells; the immunostimulatory bacteria provided herein are adenosine auxotrophs so they deplete the adenosine and can replicate in this environment.

Immune Desert Tumors

The immunostimulatory bacteria provided herein can convert or change “cold tumors” into T-cell infiltrated tumors. Tumors are classified into one of three basic immunophenotypes: immune-inflamed, immune-excluded and immune-desert phenotype (Yuan-Tong etal., (2021) Theranostics 77:5365-5386; Chen etal. (2017) Nature 547:321-3). Immune-inflamed tumors, also referred to as “hot tumors,” are characterized by high T-cell infiltration, increased interferon-y (IFN-y) signaling, expression of PD-L1, and high tumor mutational burden (TMB). Immune-excluded tumors and immune-desert tumors are “cold tumors.” In immune-excluded tumors, CD8+ T lymphocytes localize only at invasion margins and do not efficiently infiltrate the tumor; in immune-desert tumors, CD 8+ T lymphocytes are absent from the tumor and its periphery (Yuan-Tong et al., (2021) Theranostics 77:5365-5386). “Cold tumors” also are characterized by low mutational load, low major histocompatibility complex (MHC) class I expression and low PD-Ll expression (Hegde et c//. (2016) Clin Cancer 7?es.22:1865-1874). Immunosuppressive cell populations, including tumor-associated macrophages (TAMs) and T-regulatory cells (Tregs) and myeloid-derived suppressor cells (MDSCs), also are present in cold tumors.

Tumor cell death and antigen release, antigen-presenting cell (APC) processing and presentation of tumor antigens, and APC and T-cell interactions lead to T-cell priming and activation (Yuan-Tong etal. (2021) Theranostics 77:5365- 5386). Production of T cells and their physical contact with tumor cells is necessary for the success of antitumor immunity. Once infiltrating the tumor bed, cytotoxic T lymphocytes (CTLs) recognize antigenic peptide-MHC complexes on the surface of tumor cells, form immune synapses, and release perforin and granzyme to destroy the tumor cells. CTLs contribute to the apoptosis of tumor cells through the Fas/FasL pathway and suppress tumors by inducing ferroptosis and pyroptosis. Dead tumor cells release additional tumor antigens and thereby amplify the T-cell response (see, Yuan-Tong etal. (2021) Theranostics 77:5365-5386, and references cited therein).

Immune-excluded tumors and immune-desert tumors are “cold tumors.” In immune-excluded tumors, CD8+ T lymphocytes localize only at invasion margins and do not efficiently infiltrate the tumor; in immune-desert tumors, CD8+ T lymphocytes are absent from the tumor and its periphery (Yuan-Tong et al., (2021) Theranostics 11:5365-5386). “Cold tumors” also are characterized by low mutational load, low major histocompatibility complex (MHC) class I expression and low PD-L1 expression (Hegde et al. (2016) Clin Cancer Res. 22:1865-1874). Immunosuppressive cell populations, including tumor-associated macrophages (TAMs) and T-regulatory cells (Tregs) and myeloid-derived suppressor cells (MDSCs), are also present in cold tumors.

It is shown herein that treatment with therapeutics provided and described herein that result in convert the phenotype of tumor-resident macrophages into a hybrid M1/M2 phenotype, which exhibits increased phagocytosis of apoptotic cells and bacteria, can convert cold tumors into hot tumors to thereby render the tumors susceptible to treatment with immunotherapy, including checkpoint inhibitors.

In summary, among the findings described and exemplified herein, it is shown and described herein that Ml macrophages are not phagocytic of apoptotic cells; M2 macrophages are. Ml macrophages are highly inflammatory, do not produce type I IFN, and suppress CD8+ T-cell mediated adaptive immunity in humans. Infection, such as bactofection with immunostunulatory bacteria, of primary human M2 macrophages with immunostimulatory bacteria provided herein, as exemplified by the strain designated STACT IL-15plex + eSTING, induces a hybrid M1/M2 phenotype that retains M2 phagocytic capacity, upregulates Ml -like costimulatory receptors (CD80/86) and lymph node chemotaxis receptors (CCR7), and produces type I IFN- mediated cytokines and chemokines. The induction of type I IFN by the encoded payloads, such as IL-15plex (IL-15/IL-15R alpha chain complex) + eSTING, such as the constitutive STING proteins, in M2 macrophages enhances phagocytosis.

Human primary M2 macrophages, not Ml macrophages, provide for plasmid transfer and gene expression following bactofection of with a strain, such as STACT IL-15plex + eSTING. THP1 monocytes can be differentiated to macrophages using PMA, but THP1 -derived macrophages do not provide for plasmid transfer. Despite the presence of CFU in the spleen and liver following administration, and an abundance of phagocytic splenic macrophages and liver Kupffer cells, no payload expression was observed in these tissues. It is shown herein that plasmid transfer and gene expression requires proliferating cells. M2 macrophages, unlike Ml, splenic or liver Kupffer macrophages, are proliferating and therefore can provide for plasmid transfer following bactofection. Treatment of THP1 cells with PMA suppresses cell cycle genes, induces Maf-B and Maf cell cycle suppressing transcription factors, and prohibits plasmid transfer. Addition of M-CSF and co-culture with apoptotic cells enhances proliferation and plasmid transfer, while Ml -differentiating reagents and the M2 reagents IL-4 and IL- 10 impair proliferation and plasmid transfer. The ability of STACT IL-15plex + eSTING to deliver plasmid payloads in murine tumors is correlated to the amount of proliferating macrophages. Induction of type I IFN by STACT IL-15plex + eSTING enhances tumor macrophage proliferation (MMTV), and CD8 T-cell infiltration/expansion is correlated to # proliferating macrophages.

Proliferating macrophages in human solid tumors can be identified by IHC staining of CD68 + PCNA or Ki67, and qPCR for the G2M score and stathminl expression.

C. MODIFICATIONS AND ENHANCEMENTS OF IMMUNOSTIMULATORY BACTERIA TO INCREASE THERAPEUTIC INDEX AND TO INCREASE ACCUMULATION IN TUMOR-RESIDENT MYELOID CELLS

Provided herein are enhancements, including modifications to the bacterial genome, or to the immunostimulatory bacteria, that, for example, reduce toxicity and improve the anti-tumor activity, such as by increasing accumulation in tumor-resident myeloid cells, improving resistance to complement inactivation, reducing immune cell death, promoting adaptive immunity, and enhancing T-cell function. The modifications are described and exemplified with respect to Salmonella, particularly S. typhimuriunr, it is understood that the skilled person can effect similar enhancements/modifications in other bacterial species, such as Listeria, and Escherichia, and in other Salmonella strains to achieve similar properties and/or effects, and to express the same encoded payloads. Exemplary of such enhancements/modifications are the following:

1. Deletions in Genes in the LPS Biosynthetic Pathway

The lipopolysaccharide (LPS) of Gram-negative bacteria is the major component of the outer leaflet of the bacterial membrane. It is composed of three major parts, lipid A, a non-repeating core oligosaccharide, and the O-antigen (or O polysaccharide). O-antigen is the outermost portion on LPS and serves as a protective layer against bacterial permeability, however, the sugar composition of O-antigen varies widely between strains. The lipid A and core oligosaccharide vary less, and are more typically conserved within strains of the same species. Lipid A is the portion of LPS that contains endotoxin activity. It is typically a disaccharide decorated with multiple fatty acids. These hydrophobic fatty acid chains anchor the LPS into the bacterial membrane, and the rest of the LPS projects from the cell surface. The lipid A domain is responsible for much of the toxicity of Gram-negative bacteria. Typically, LPS in the blood is recognized as a significant pathogen associated molecular pattern (PAMP), and induces a profound pro-inflammatory response. LPS is the ligand for a membrane-bound receptor complex comprising CD14, MD2, and TLR4. TLR4 is a transmembrane protein that can signal through the MyD88 and TRIF pathways to stimulate the NF-KB pathway and result in the production of pro-inflammatory cytokines, such as TNF-a and IL-6, the result of which can be endotoxic shock, which can be fatal. LPS in the cytosol of mammalian cells can bind directly to the caspase recruitment domains (CARDs) of caspases 4, 5, and 11, leading to autoactivation and pyroptotic cell death (see, e.g., Hagar et al. (2015) Cell Research 25: 149-150). The composition of lipid A and the toxigenicity of lipid A variants is well documented. For example, a monophosphorylated lipid A is much less inflammatory than lipid A with multiple phosphate groups. The number and length of the acyl chains on lipid A also can have a profound impact on the degree of toxicity. Canonical lipid A from E. coli has six acyl chains, and this hexa-acyl ati on is potently toxic. A typhimurium lipid A is similar to that of E. coli,' it is a glucosamine disaccharide that carries four primary and two secondary hydroxyacyl chains (see, e.g., Raetz et al. (2002) Amu. Rev. Biochem. 71:635-700). a. tnsbB Deletion

The enzyme lipid A biosynthesis myristoyltransferase, encoded by the msbB gene in A typhimurium, catalyzes the addition of a terminal myristoyl group to the lipid A domain of lipopolysaccharide (LPS) (see, e.g., Low et al. (1999) Nat. Biotechnol. 17(l):37-41). Deletion of msbB, thus, alters the acyl composition of the lipid A domain of LPS, the major component of the outer membranes of Gramnegative bacteria. For example, deletion of msbB in the S. typhimurium strain VNP20009 results in the production of a predominantly penta-acylated lipid A, which is less toxic than native hexa-acylated lipid A, and allows for systemic delivery without the induction of toxic shock (see, e.g., Lee et al. (2000) International Journal of Toxicology 19:19-25). This modification significantly reduces the ability of the LPS to induce septic shock, attenuating the bacterial strain, and thus, increasing the therapeutic index of A^/mone/Za-based immunotherapeutics (see, e.g., U.S. Patent Publication Nos. 2003/0170276, 2003/0109026, 2004/0229338, 2005/0255088, and 2007/0298012). Importantly, msbB mutants that do not express the msbB product are unable to replicate intracellularly, as exemplified herein (see, e.g., Example 2), which is a requirement for Salmonella virulence (see, e.g., Leung etal. (1991) Proc. Natl. Acad. Sci. U.S.A. 88: 11470-11474).

Other LPS mutations, including replacements, deletions, or insertions that alter LPS expression, can be introduced into the bacterial strains provided herein, including the Salmonella strains, that dramatically reduce virulence, and thereby provide for lower toxicity, and permit the administration of higher doses. As exemplified herein, the msbB' locus can be partially deleted, or interrupted, or translocated. It also can be completely deleted, which can improve growth of the strain.

Corresponding genes, encoding homologs or orthologs of lipid A biosynthesis myristoyltransferase in other bacterial species, also can be deleted or disrupted to achieve similar results. These genes include, but are not limited to, for example, IpxM, encoding myristoyl-acyl carrier protein-dependent acyltransferase in E. coir, and msbB, encoding lipid A acyltransferase in S. typhi. b. pagP Deletion or inactivation As described above, msbB mutants of . typhimurium cannot undergo the terminal myristoylation of lipid A, and produce predominantly penta-acyl ated lipid A that is significantly less toxic than hexa-acylated lipid A. The modification of lipid A with palmitate is catalyzed by the enzyme lipid A palmitoyltransferase (PagP). Transcription of the pagP gene is under control of the PhoP/PhoQ system, which is activated by low concentrations of magnesium, e.g., inside the SCV. Thus, the acyl content of . typhimurium lipid A is variable, and with wild-type bacteria, it can be hexa-acylated or penta-acylated. The ability of S. typhimurium to palmitate its lipid A increases resistance to antimicrobial peptides that are secreted into phagolysosomes.

In wild-type S. typhimurium, expression of pagP results in lipid A that is hepta-acylated. In an msbB mutant (in which the terminal acyl chain of the lipid A cannot be added), the induction of pagP results in a hexa-acylated lipid A (see, e.g., Kong etal. (2011) Infection andlmmunity 79(12):5027-5038). Hexa-acylated lipid A has been shown to be the most pro-inflammatory. While groups have sought to exploit this pro-inflammatory signal, for example, by deletion or disruption of pagP to allow only hexa-acylated lipid A to be produced (see, e.g., Feigner etal. (2016) Gut Microbes 7(2): 171-177; and Feigner et l. (2018) Oncoimmunology 7(2):el382791), this can lead to poor tolerability, due to the TNF-a-mediated pro-inflammatory nature of the LPS, and paradoxically less adaptive immunity (see, e.g., Kocijancic et al. (2017) Oncotarget 8(30):49988-50001).

LPS is a potent TLR4 agonist that induces TNF-a and IL-6. The dose-limiting toxi cities in the I.V. VNP20009 clinical trial (see, e.g., Toso et al. (2002) J. Clin. Oncol. 20(1): 142-152) at 1E9 CFUs/m ² , were cytokine mediated (fever, hypotension), with TNF-a levels > 100,000 pg/ml, and IL-6 levels > 10,000 pg/ml in serum at 2 hours. Despite the msbB deletion in VNP20009 and its reduced pyrogenicity, the LPS still can be toxic at high doses, possibly due to the presence of hexa-acylated lipid A. Thus, a pagP'/msbB' strain, which cannot produce hexa-acylated lipid A, and produces only penta-acylated lipid A, resulting in lower induction of pro- inflammatory cytokines, is better tolerated at higher doses, and will allow for dosing in humans at or above 1E9 CFUs/m ² . Higher dosing leads to increased colonization of tumors, tumor-resident immune cells, and the tumor microenvironment, enhancing the therapeutic efficacy of the immunostimulatory bacteria. Because of the resulting change in bacterial membranes and structure, the host immune response, such as complement activity, is altered so that the bacteria are not eliminated upon systemic administration. For example, it is shown herein that pagP /msbB' mutant strains have increased resistance to complement inactivation and enhanced stability in human serum.

Provided herein are immunostimulatory bacteria, exemplified by live attenuated Salmonella strains, such as the exemplary strain of S. typhimurium, that only can produce LPS with penta-acyl ated lipid A, that contain a deletion or disruption of the msbB gene, and that further are modified by deletion or disruption of pagP. As discussed above, deletion of msbB expression prevents the terminal myristoylation of lipid A, while deletion of pagP expression prevents palmitoylation. A strain modified to produce LPS penta-acylated lipid A results in lower levels of pro-inflammatory cytokines, improved stability in the blood, resistance to complement fixation, increased sensitivity to antimicrobial peptides, enhanced tolerability, and increased anti-tumor immunity when further modified to express heterologous genetic payloads that stimulate the immune response in the tumor microenvironment.

Corresponding genes, encoding homologs and orthologs of lipid A palmitoyltransferase (PagP) in other bacterial species, also can be deleted or disrupted to achieve similar results. These genes include, but are not limited to, for example, pagP, encoding Lipid IVA palmitoyltransferase in E. coir, and pagP, encoding antimicrobial peptide resistance and lipid A acylation protein in S. typhi.

2. Nutrient Auxotrophy

The immunostimulatory bacteria provided herein can be attenuated by rendering them auxotrophic for one or more essential nutrients, such as purines (for example, adenine), nucleosides (for example, adenosine), amino acids (for example, aromatic amino acids, arginine, and leucine), adenosine triphosphate (ATP), or other nutrients as known and described in the art. a. purl Deletion/Disruption

Phosphoribosylaminoimidazole synthetase, an enzyme encoded by the purl gene (synonymous with the purM gene), is involved in the biosynthesis pathway of purines. Disruption or deletion or inactivation of the purl gene, thus, renders the bacteria auxotrophic for purines. In addition to being attenuated, purl' mutants are enriched in the tumor environment and have significant anti -tumor activity (see, e.g., Pawelek et al. (1997) Cancer Research 57:4537-4544). It was previously described that this colonization results from the high concentration of purines present in the interstitial fluid of tumors as a result of their rapid cellular turnover. Since the purl' bacteria are unable to synthesize purines, they require an external source of adenine, and it was thought that this would lead to their restricted growth in the purine- enriched tumor microenvironment (see, e.g., Rosenberg et al. (2002) J. Immunotherapy 25(3):218-225). While the VNP20009 strain was initially reported to contain a deletion of the purl gene (see, e.g., Low etal. (2003) Methods in Molecular Medicine Vol. 90, Suicide Gene Therapy. Methods and Reviews, pp. 47-59), subsequent analysis of the entire genome of VNP20009 demonstrated that the purl gene is not deleted, but is disrupted by a chromosomal inversion (see, e.g., Broadway et al. (2014) Journal of Biotechnology 192:177-178). The entire gene is contained within two parts of the VNP20009 chromosome that is flanked by insertion sequences, one of which has an active transposase. While disruption of the purl gene limits replication to the tumor tissue/microenvironment, it still permits intracellular replication and virulence. Deletion or disruption of each of the msbB and the purl genes, as exemplified herein (see, Example 2), is required to limit bacterial growth to the extracellular space in tumor tissue, and prevent intracellular replication. Provided herein are strains in which the coding portion of these genes are completely deleted to eliminate any possible reversion to wild-type by recombination. It is shown herein that such bacteria grow more effectively.

Besides purl gene deletions or disruptions, nutrient auxotrophy can be introduced into the immunostimulatory bacteria by deletions/mutations in genes such as aro, gua, thy, nad, and asd, for example. Nutrients produced by the biosynthesis pathways involving these genes are often unavailable in host cells, and as such, bacterial survival is challenging. For example, attenuation of Salmonella and other bacterial species can be achieved by deletion of the aroA gene, which is part of the shikimate pathway, connecting glycolysis to aromatic amino acid biosynthesis (see, e.g., Feigner et al. (2016) mBio 7(5):e01220-16). Deletion of aroA results in bacterial auxotrophy for aromatic amino acids and subsequent attenuation (see, e.g., U.S. Patent Publication Nos. 2003/0170276, 2003/0175297, 2012/0009153, and 2016/0369282; and International Application Publication Nos. WO 2015/032165 and WO 2016/025582). Similarly, other enzymes involved in the biosynthesis pathway for aromatic amino acids, including aroC and aroD, have been deleted to achieve attenuation (see, e.g., U.S. Patent Publication No. 2016/0369282; and International Application Publication No. WO 2016/025582). For example, S. typhimurium strain SL7207 is an aromatic amino acid auxotroph (aroA' mutant); strains Al and Al-R are leucine-arginine auxotrophs; and VNP20009/YS1646 is a purine auxotroph (purl' mutant) as well as being msbBP As shown herein, VNP20009/YS1646 is also auxotrophic for the immunosuppressive nucleoside adenosine, and for ATP (see, e.g., Example 1). Strains provided herein include strains derived from the strain designated YS1646, such as those that lack flagella, are pagP~ or modified to produce pentaacylated LPS, and include additional modifications, including complete deletion of purl and/msbB, as well as deletion of the curli fimbriae, such as by genome modifications that render the bacterium csglp and additional modifications that require various nutrients for growth, such as thyA~ strains. The strains also can have genome modifications that render them ansB' so that they do not produce asparagine synthase, which can inhibit T cells, thereby eliminating this immunosuppressive aspect of immunostimulatory bacteria. Exemplary strains include those designated Y S I 646 < .«// EELGIApagP/AansB/AcsgD/F-Apurl, and

Y S I 646 tf.sz// M GIApagP AansB, Acsg!) F-ApurL 'AthyA. It is understood that these designations reference Salmonella genes; similar modifications can be effected in other bacterial species, such as Listeria, and E. coli, such as Nissle.

Corresponding genes, encoding homologs or orthologs of phosphoribosylaminoimidazole synthetase (Purl), and other genes required for purine synthesis in other bacterial species, also can be deleted or disrupted to achieve similar results. These genes include, but are not limited to, for example, purM, encoding phosphoribosylformylglycinamide cyclo-ligase in E. coli,' purM, encoding phosphoribosylformylglycinamidine cyclo-ligase in S. typhv, purA, encoding adenylosuccinate synthetase, purQ, encoding phosphoribosylformylglycinamidine synthase II, and purS, encoding phosphoribosylformylglycinamidine synthase subunit PurS in A. monocytogenes,' purM (BL 1122), encoding phosphoribosylformylglycinamidine cyclo-ligase in Bifidobacterium longunr, and NT01CX_RS09765, encoding AIR synthase, and NT01CX_RS07625 (purM), encoding phosphoribosylformylglycinamidine cyclo-ligase in Clostridium novyi. b. Adenosine Auxotrophy

Metabolites derived from the tryptophan and adenosine triphosphate (ATP)/ adenosine pathways are major drivers in forming an immunosuppressive environment within the tumor/tumor microenvironment (TME). Adenosine, which exists in the free form inside and outside of cells, is an effector of immune function. Adenosine decreases T-cell receptor induced activation OI NF-KB, and inhibits IL-2, IL-4, and IFN-y. Adenosine decreases T-cell cytotoxicity, increases T-cell anergy, and increases T-cell differentiation to FOXP3 ⁺ or LAG3 ⁺ regulatory T-cells (T-reg cells, T-regs or Tregs). On natural killer (NK) cells, adenosine decreases IFN-y production, and suppresses NK cell cytotoxicity. Adenosine blocks neutrophil adhesion and extravasation, decreases phagocytosis, and attenuates levels of superoxide and nitric oxide. Adenosine also decreases the expression of TNF-a, IL-12, and MIP-la (CCL3) on macrophages, attenuates major histocompatibility complex (MHC) Class II expression, and increases levels of IL-10 and IL-6. Adenosine immunomodulation activity occurs after its release into the extracellular space of the tumor and activation of adenosine receptors (ADRs) on the surface of target immune cells, cancer cells or endothelial cells. The high adenosine levels in the tumor microenvironment result in local immunosuppression, which limits the capacity of the immune system to eliminate cancer cells.

Extracellular adenosine is produced by the sequential activities of membrane associated ectoenzymes, CD39 (ecto-nucleoside triphosphate diphosphohydrolase 1, or E-NTPDasel) and CD73 (ecto-5’ -nucleotidase), which are expressed on tumor stromal cells, together producing adenosine by phosphohydrolysis of ATP or ADP produced from dead or dying cells. CD39 converts extracellular ATP (or ADP) to 5’- AMP, which is converted to adenosine by CD73. Expression of CD39 and CD73 on endothelial cells is increased under the hypoxic conditions of the tumor microenvironment, thereby increasing levels of adenosine. Tumor hypoxia can result from inadequate blood supply and disorganized tumor vasculature, impairing delivery of oxygen (see, e.g., Carroll and Ashcroft (2005) Expert. Rev. Mol. Med. 7(6), DOI: 10.1017/S1462399405009117). Hypoxia, which occurs in the tumor microenvironment, also inhibits adenylate kinase (AK), which converts adenosine to AMP, leading to very high extracellular adenosine concentrations. The extracellular concentration of adenosine in the hypoxic tumor microenvironment has been measured at 10-100 pM, which is up to about 100-1000 fold higher than the typical extracellular adenosine concentration of approximately 0.1 pM (see, e.g., Vaupel et al. (2016) Adv. Exp. Med. Biol. 876:177-183; and Antonioli etal. 2Q ) Nat. Rev. Can. 13:842-857). Since hypoxic regions in tumors are distal from microvessels, the local concentration of adenosine in some regions of the tumor can be higher than in others. To direct effects to inhibit the immune system, adenosine also can control cancer cell growth and dissemination by effects on cancer cell proliferation, apoptosis and angiogenesis. For example, adenosine can promote angiogenesis, primarily through the stimulation of AZA and AZB receptors. Stimulation of the receptors on endothelial cells can regulate the expression of intercellular adhesion molecule 1 (ICAM-1) and E-selectin on endothelial cells, maintain vascular integrity, and promote vessel growth (see, e.g., Antonioli etal. (2013) Nat. Rev. Can. 13:842-857). Activation of one or more of AZA, A B, or A3 on various cells by adenosine can stimulate the production of the pro-angiogenic factors, such as vascular endothelial growth factor (VEGF), interleukin-8 (IL-8) or angiopoietin 2 (see, e. ., Antonioli et al. (2013) Nat. Rev. Can. 13:842-857).

Adenosine also can directly regulate tumor cell proliferation, apoptosis, and metastasis through interaction with receptors on cancer cells. For example, studies have shown that the activation of Ai and AZA receptors promote tumor cell proliferation in some breast cancer cell lines, and activation of AZB receptors have cancer growth-promoting properties in colon carcinoma cells (see, e.g., Antonioli et al. (2013) Nat. Rev. Can. 13:842-857). Adenosine also can trigger apoptosis of cancer cells, and various studies have correlated this activity to activation of the extrinsic apoptotic pathway through A3, or the intrinsic apoptotic pathway through AZA and AZB (see, e.g., Antonioli etal. (2013)). Adenosine can promote tumor cell migration and metastasis, by increasing cell motility, adhesion to the extracellular matrix, and expression of cell attachment proteins and receptors to promote cell movement and motility.

The extracellular release of adenosine triphosphate (ATP) occurs from stimulated immune cells, and from damaged, dying, or stressed cells. The NLR family pyrin domain-containing 3 (NLRP3) inflammasome, when stimulated by this extracellular release of ATP, activates caspase- 1 and results in the secretion of the cytokines IL-ip and IL-18, which in turn activate innate and adaptive immune responses (see, e.g., Stagg and Smyth (2010) Oncogene 29:5346-5358). ATP can accumulate to concentrations exceeding 100 mM in tumor tissue, whereas levels of ATP found in healthy tissues are very low (~l-5 pM) (see, e.g., Song et al. (2016) Am. J. Physiol. Cell Physiol. 310(2):C99-C114). ATP is catabolized into adenosine by the enzymes CD39 and CD73. Activated adenosine acts as a highly immunosuppressive metabolite via a negative-feedback mechanism and has a pleiotropic effect against multiple immune cell types in the hypoxic tumor microenvironment (see, e.g., Stagg and Smyth (2010) Oncogene 29:5346-5358). Adenosine receptors A A and AZB are expressed on a variety of immune cells and are stimulated by adenosine to promote cAMP -mediated signaling changes, resulting in immunosuppressive phenotypes of T- cells, B-cells, NK cells, dendritic cells (DCs), mast cells, macrophages, neutrophils, and natural killer T (NKT) cells. As a result, adenosine levels can accumulate to over one hundred times their normal concentration in pathological tissues, such as solid tumors, which have been shown to overexpress ecto-nucleotidases, such as CD73. Adenosine also has been shown to promote tumor angiogenesis and development. An engineered bacterium that is auxotrophic for adenosine would thus exhibit enhanced tumor-targeting and colonization.

Immunostimulatory bacteria, such as Salmonella iyphi, can be made auxotrophic for adenosine, for example, by deletion of the tsx gene (see, e.g., Bucarey et al. (2005) Infection and Immunity 73(10):6210-6219) or by deletion of purD (see, e.g., Husseiny (2005) Infection and Immunity 73(3): 1598-1605). In the Gramnegative bacteria Xanthomonas oryzae, a purD gene knockout was shown to be auxotrophic for adenosine (see, e.g., Park et al. (2007) FEMS Microbiol. Lett. 276:55- 59). As exemplified herein, A typhimurium strain VNP20009 is auxotrophic for adenosine due to its purl modification; hence, further modification to render it auxotrophic for adenosine is not required. Hence, embodiments of the immunostimulatory bacterial strains, as provided herein, are auxotrophic for adenosine. Such auxotrophic bacteria selectively replicate in the tumor microenvironment, further increasing accumulation and replication of the administered bacteria in tumors, and decreasing the levels of adenosine in and around tumors, thereby reducing or eliminating the immunosuppression caused by the accumulation of adenosine. Exemplary of such bacteria, provided herein, is a modified strain of A typhimurium containing purL/msbB~ mutations to provide adenosine auxotrophy. For other strains and bacteria, the purl gene can be disrupted as it has been in VNP20009, or it can contain a deletion of all or a portion of the purl gene, which ensures that there cannot be a reversion to a wild-type gene. As described elsewhere herein, in strain VNP20009, the purl gene was inactivated by inversion. Similarly, the msbB gene in VNP20009 was not completely deleted. As exemplified herein, strains in which the purl and msbB genes have been completely deleted to eliminate any risk of reversion, demonstrate superior fitness as assessed by growth of cultures in vitro.

Immunostimulatory bacteria modified by rendering them auxotrophic for one or more essential nutrients, such as purines (for example, adenine), nucleosides (for example, adenosine), amino acids (for example, aromatic amino acids, arginine, and leucine), or adenosine triphosphate (ATP), are employed. In particular, in embodiments of the immunostimulatory bacteria provided herein, such as strains of S. typhimurium, the bacteria are rendered auxotrophic for adenosine, and optionally, for ATP, and preferentially accumulate in tumor microenvironments (TMEs). Hence, strains of immunostimulatory bacteria described herein are attenuated because they require purines, adenosine, and/or ATP for growth, and they preferentially colonize TMEs, which, as discussed below, have an abundance of these metabolites. Because adenosine accumulation that occurs in the tumor microenvironment of some tumors is immunosuppressive, adenosine auxotrophy eliminates the immunosuppression from adenosine that accumulates in the tumor microenvironment. c. Thymidine Auxotrophy

Genome modifications can be introduced in place of or in addition to the inactivation/deletion (see section 3) discussed below. Other deletions or inactivation of genes or gene products required for growth, such as genes that produce nutrients, can be used in place of or in addition to, for example, the asd inactivation/deletion. These include, for example, modifications that render the bacteria thyA- (see, e.g., Loessner et al. (2006) FEBS Lett 265:81-88). Immunostimulatory bacteria that are thyA~ have genome modifications, such as insertions, deletions, replacements, transpositions, and/or other changes, that result in inactive or eliminate production of thymidylate synthase. Thymidylate synthase catalyzes the reductive methylation of dUMP to dTMP, a DNA biosynthesis precursor (precursor to dTTP).

Elimination of expression or production or other attenuating mutations of the bacterial genome for production of such products results in release of encoded macromolecules upon bacterial cell death in vivo after administration. Asd, as discussed below, is an essential enzyme for bacterial cell wall synthesis; ThyA is an enzyme needed for DNA synthesis. Mutation of the respective genes renders the strain auxotrophic for diaminopimelic acid (DAP) or thymidine monophosphate precursors. Upon deprivation of the complementing substrates, such bacteria die by DAP -less or thymine-less death, resulting in release of bacterial proteins and plasmid. Inactivation or elimination of Asd, results in release of macromolecules. Elimination or inactivation of Thy A (to produce A/AjA bacteria, which includes those with insertions, deletions, and other modifications so that active enzyme is not produced) expression/activity does not result in release of macromolecules, including proteins and plasmid, upon thymidine starvation (Loessner et al., (2006) FEBS Lett 265 :81- 88).

Thus, EthyA bacteria, in which the genome is modified so that active enzyme is not produced, are advantageous, for example, for in vivo delivery of plasmids to host cells, since the bacteria do not prematurely release their contents. Since the bacteria provided herein infect or accumulate in myeloid cells, such as phagocytic cells, such as macrophages, dendritic cells, monocytes, and neutrophils, which consume bacteria, the intact EthyA bacteria, release the plasmid encoding the payloads, such as therapeutic products, inside the targeted cells for, for example, expression, ifRNA, or secretion or presentation, if protein. Thus, genome modifications that render the bacteria thyA~ have advantages for particular applications, such as immunization, presentation on cells, delivery of RNA, and other such applications.

3. Plasmid Maintenance and Delivery a. asd Deletion

As discussed above, the bacteria can be rendered thyA',- they also can be rendered asd instead or in addition. The selection of the particular genome modification depends upon the intended application of the bacteria. thyA inactivation is advantageous for some applications; for others rendering the bacteria asd is advantageous; it is within the skill of the skilled person to select the particular modification(s). The asd gene in bacteria encodes an aspartate-semialdehyde dehydrogenase, asd mutants of S. typhimurium have an obligate requirement for diaminopimelic acid (DAP), which is required for cell wall synthesis, and will undergo lysis in environments deprived of DAP. This DAP auxotrophy can be used for plasmid selection and maintenance of plasmid stability in vivo, without the use of antibiotics, when the asd gene is complemented in trans on a plasmid in the bacterium. Non-antibiotic-based plasmid selection systems are advantageous, and allow for 1) the use of administered antibiotics as a rapid clearance mechanism in the event of adverse symptoms, and 2) for antibiotic-free scale up of production, where such use is commonly avoided. The asd gene complementation system provides for such non-antibiotic-based plasmid selection (see, e.g., Galan et al. (1990) Gene 94(l):29-35). The use of the asd gene complementation system to maintain plasmids in the tumor microenvironment is expected to increase the potency of S. typhimurium strains engineered to deliver plasmids encoding genetic payloads/therapeutic products, such as immunostimulatory proteins (e.g., cytokines, chemokines, costimulatory molecules); cytosolic DNA/RNA sensors that induce type I IFN, such as STING and IRF3, and gain-of-function/constitutively active mutants thereof; antibodies and fragments thereof (e.g., checkpoint inhibitors, or anti -IL-6 or anti- VEGF antibodies); bi-specific T-cell engagers (sold under the trademark BiTEs®); interfering RNAs; and other therapeutic products as discussed elsewhere herein and known in the art; and complementary combinations of all of the preceding therapeutic products.

An alternative use for an asd mutant of S. typhimurium is to exploit the DAP auxotrophy to produce an autolytic (or suicidal) strain, for delivery of therapeutic products/macromolecules to infected cells, without the ability to persistently colonize host tumors. Deletion of the asd gene makes the bacteria auxotrophic for DAP when grown in vitro or in vivo. An example described herein, provides an asd deletion strain that is auxotrophic for DAP, and that contains a plasmid suitable for delivery of immunomodulatory proteins, but that does not contain an asd complementing gene, resulting in a strain that is defective for replication in vivo. This strain is propagated in vitro in the presence of DAP, and grows normally, and then is administered as an immunotherapeutic agent to a mammalian host where DAP is not present. The suicidal strain is able to invade host cells but is not able to replicate due to the absence of DAP in mammalian tissues, lysing automatically, and delivering its cytosolic contents (e.g., plasmids or proteins).

Corresponding genes, encoding homologs or orthologs of aspartatesemialdehyde dehydrogenase (asd) in other bacterial species, also can be deleted or disrupted to achieve similar results. These genes include, but are not limited to, for example, asd, encoding aspartate-semialdehyde dehydrogenase in E. coli asd (STY4271), encoding aspartate-semialdehyde dehydrogenase in S. typhi, asd (lmol437), encoding aspartate-semialdehyde dehydrogenase in A. monocytogenes,' asd (BL0492), encoding aspartate-semialdehyde dehydrogenase in Bifidobacterium longunr, and NT01CX_RS04325 (asd), encoding aspartate-semialdehyde dehydrogenase in Clostridium novyi. Similarly, as discussed, above, thyA can be inactivated or eliminated by genome modification in place of the asd modification. As discussed above, inactivation or elimination of thymidylate synthase expression/activity does not result in release of macromolecules, including proteins and plasmids, upon thymidine starvation. b. endA Deletion/Disruption

The endA gene (see, for example, SEQ ID NO:250) encodes an endonuclease (DNA-specific endonuclease I; see, for example, SEQ ID NO:251) that mediates degradation of double-stranded DNA (dsDNA) in the periplasm of Gram-negative bacteria. Most common strains of laboratory E. coli are endA', as a mutation in the endA gene allows for higher yields of plasmid DNA. This gene is conserved among species. To facilitate intact plasmid DNA delivery, the endA gene of the engineered immunostimulatory bacteria is deleted or mutated to prevent its endonuclease activity. Exemplary of such mutations is an E208K amino acid substitution (see, e.g., Durfee el al. (2008) J. Bacterial. 190(7):2597-2606), or a corresponding mutation in the species of interest. endA, including residue E208, is conserved among bacterial species, including Salmonella. Thus, the E208K mutation can be used to eliminate endonuclease activity in other species, including Salmonella species. Those of skill in the art can introduce other mutations or deletions to eliminate endA activity. Effecting this mutation, or deleting or disrupting the gene to eliminate activity of endA in the immunostimulatory bacteria herein, such as in Salmonella, increases efficiency of intact plasmid DNA delivery, thereby increasing expression of any one, or two, or more, immunomodulatory proteins/therapeutic products encoded on the plasmid, and enhancing the anti-tumor immune response and anti-tumor efficacy.

4. Flagellin Knockout Strains

Flagella are organelles on the surface of bacteria that are composed of a long filament that is attached, via a hook, to a rotary motor that can rotate in a clockwise or counterclockwise manner to provide a means for locomotion. Flagella, for example, in S. typhimurium, are important for chemotaxis and for establishing an infection via the oral route, due to the ability to mediate motility across the mucous layer in the gastrointestinal tract. While flagella have been demonstrated to be required for chemotaxis to and colonization of tumor cylindroids in vitro (see, e.g., Kasinskas and Forbes (2007) Cancer Res. 67(7):3201-3209), and motility has been shown to be important for tumor penetration (see, e.g., Toley and Forbes (2012) Integr. Biol. (Camb) 4(2): 165-176), flagella are not required for tumor colonization in animals when the bacteria are administered intravenously (see, e.g., Stntzker et al. (2010) International Journal of Medical Microbiology 300:449-456). Each flagellar filament is composed of tens of thousands of flagellin subunits. The 5. typhimurium chromosome contains two genes, fliC and fljB, that encode antigenically distinct flagellin monomers. Mutants defective for both fliC and fljB are nonmotile and avimlent when administered via the oral route of infection, but maintain virulence when administered parenterally.

Flagellin is a major pro-inflammatory determinant of Salmonella (see, e.g., Zeng et al. (2003) J. Immunol. 171 :3668-3674), and is directly recognized by TLR5 on the surface of cells, and by NLCR4 in the cytosol (see, e.g., Lightfield et al. (2008) Nat. Immunol. 9(10): 1171-1178). Both pathways lead to pro-inflammatory responses resulting in the secretion of cytokines, including IL- Ip, IL- 18, TNF-a, and IL-6. Attempts have been made to make 5a/ o«e//«-based cancer immunotherapy more potent by increasing the pro-inflammatory response to flagellin by engineering the bacteria to secrete Vibrio vulnificus flagellin B, which induces greater inflammation than flagellin encoded by fliC and fljB (see, e.g., Zheng et al. (2017) Sci. Transl. Med. 9(376):eaak9537).

Herein, Salmonella bacteria, such as A typhimurium, are engineered to lack both flagellin subunits fliC and fljB, to reduce TLR5-mediated pro-inflammatory signaling. Other bacteria that contain flagella similarly can be engineered to eliminate flagella, and to include other modifications that have the substantially the same effects as the exemplified modifications of Salmonella. For example, as shown herein, a Salmonella strain lacking msbB and/or pagP, which results in reduced TNF-alpha induction, is combined with elimination of flagella, which can be achieved in Salmonella by fliC and fljB knockouts. This results in a Salmonella strain that has a combined reduction in, among other effects, TNF-alpha induction and a reduction in TLR5 recognition. These bacterial modifications, msbB' , pagP~ , fliC’ , and fljB', can be combined with an immunostimulatory plasmid, optionally containing CpGs, encoding therapeutic products, such as immunomodulatory proteins, and combinations of products. The resulting bacteria have reduced pro-inflammatory signaling, but robust anti-tumor activity. These genome modifications can be combined with others of the genome modifications described herein as well. For vaccines, the encoded products can include antigens and proteins against which an immmunoprotective or therapeutic response is intended. For example, as exemplified and provided herein, a fliC and fljB double mutant was constructed in the c -dcletcd strain of 5. typhimurium, VNP20009. VNP20009, which is attenuated for virulence by disruption of purUpurM, and which contains a modification of the msbB gene (a partial deletion) that results the in production of a lipid A subunit that is less toxigenic than wild-type lipid A. This results in reduced TNF-a production in a mouse model after intravenous administration, compared to strains with wild-type lipid A. The resulting strain is exemplary of strains that are attenuated for bacterial inflammation by modification of lipid A to reduce TLR2/4 signaling, and by deletion of expression of the flagellin subunits to reduce TLR5 recognition and inflammasome induction.

Pathogenesis in certain bacterial species, including Salmonella species, such as S. typhimurium, involves a cluster of genes referred to as Salmonella pathogenicity islands (SPIs). Salmonella invades non-phagocytic intestinal epithelial cells using a type 3 secretion system (T3SS) encoded by the Salmonella pathogenicity island 1 (SPI-1), which forms a needle-like structure that injects effector proteins directly into the cytosol of host cells. These effector proteins lead to rearrangement of the eukaryotic cell cytoskeleton to facilitate invasion of the intestinal epithelium, and also induces proinflammatory cytokines. The SPI designated SPI-1 mediates invasion of epithelial cells. SPI-1 genes include, but are not limited to: avrA, hilA, hilD, invA, invB, invC, invE, invF, invG, invH, invl, invJ, iacP, iagB, spaO, spaP, spaQ, spaR, spaS, orgA, orgB, orgC,prgH,prgI, prgj, prgK, sicA, sicP, sipA, sipB, sipC, sipD, sirC, sopB, sopD, sopE, sopE2, sprB, and sptP. Deletion of one or more of these genes reduces or eliminates the ability of the bacterium to infect epithelial cells, but does not affect their ability to infect or invade phagocytic cells, including phagocytic immune cells. For example, it was demonstrated that deletion of both the fliC and fljB genes significantly reduced expression of SPI-1 genes, such as hilA, hilD, invA, invF, and sopB, thereby reducing the ability to invade non-phagocytic cells (see, e.g., Elhadad eZ a/. (2015) Infect. Immun. 83(9):3355-3368).

In bacteria such as Salmonella, flagellin, in addition to the SPI-1 type 3 secretion system (T3SS), is necessary for triggering pyroptosis in macrophages, and can be detected by the macrophage NLRC4 inflammasome. Elimination of flagellin subunits decreases pyroptosis in macrophages. For example, S. typhimurium with deletions in fliC and fljB results in significantly reduced IL-10 secretion compared to the wild-type strain, whereas cellular uptake and intracellular replication of the bacterium remains unaffected. This demonstrates that flagellin plays a significant role in inflammasome activation. Additionally, 5. typhimurium strains engineered to constitutively express fliC were found to induce macrophage pyroptosis (see, e.g., Li et al. (2016) Scientific Reports 6:37447; Fink and Cookson (2007) Cellular Microbiology 9(l l):2562-2570; and Winter etal. (2015) Infect. Immun. 83(4): 1546- 1555).

The genome of the immunostimulatory bacteria herein can be modified to delete or mutate the flagellin genes fliC and fljB in S. typhimurium, leading to decreased cell death of tumor-resident immune cells, such as macrophages, and enhancing the anti-tumor immune response of the immunostimulatory bacteria.

Deletion of the flagellin subunits, combined with modification of the LPS, allows for greater tolerability in the host, limits uptake into only phagocytic cells and decreases their pyroptotic cell death, and directs the immunostimulatory response towards delivery of therapeutic products, such as immunomodulatory proteins, to the TME, particularly tumor-resident myeloid cells. The resulting immunostimulatory bacteria elicit an anti-tumor response, and promote an adaptive immune response to the tumor.

Corresponding genes, encoding flagellin in other bacterial species, also can be deleted to achieve similar results. These genes include, but are not limited to, for example, fliC, encoding flagellar filament structural protein, and fliE, encoding flagellar basal -body protein FliE in A. colyfliC, encoding flagellin and flgB, encoding flagellar basal -body rod protein FlgB, in S. typhi,flaA encoding flagellin, fliE, encoding flagellar hook-basal body protein FliE, and flgB, encoding flagellar basal- body rod protein FlgB, in L. monocytogenes, and NT01CX_RS04995, NT01CX_RS04990, NT01CX_RS05070 and NT01CX_RS05075, encoding flagellin, NT01CX_RS05080 (flgB , encoding flagellar basal body rod protein FlgB, NT01CX_RS05085 (flgC), encoding flagellar basal body rod protein FlgC, and NT01CX_RS05215 (flgG), encoding flagellar basal body rod protein FlgG, in Clostridium novyi.

5. Engineering Bacteria to Promote Adaptive Immunity and Enhance T-Cell Function

L-Asparaginase II (ansB) Deletion/Disruption

L-asparaginase II is an enzyme that catalyzes conversion of L-asparagine to ammonia and aspartic acid. Several bacterial strains, such as E. coli and S. typhimurium, utilize L-asparaginase to scavenge fructose-asparagine as a carbon and nitrogen source (see, e.g., Sabag-Daigle et al. (2018) ^4/?/?/. Environ. Microbiol. 84(5):e01957-17). Malignant T-cells, such as in acute lymphoblastic leukemia (ALL), require asparagine as they lack the enzymes to synthesize it. Administration of L- asparaginases has been a frontline therapy for ALL since the early 1970’s (see, e.g., Batool et al. (2016) Appt. Biochem. Biotechnol. 178(5):900-923). Production of L- asparaginase II by 5. typhimurium is both necessary and sufficient for T-cell inhibition, as it directly induces T-cell receptor (TCR) downregulation, decreases T- cell cytokine production, and inhibits tumor cytolytic function (see, e.g., Kullas et al. (2012) Cell Host Microbe 12(6)791-798; and van der Velden et al. (2005) Proc. Natl. Acad. Sci. U.S.A. 102(49): 17769-17774). Under rapid clonal expansion conditions, such as those that occur during T-cell activation in the tumor microenvironment, asparagine is required, and its depletion by L-asparaginase II leads to T-cell suppression. L-asparaginase II, thus, has been used as an anti-cancer therapeutic for cancers in which T-cell suppression is a therapeutic modality.

In contrast to the prior uses of L-asparaginase as an anti-cancer therapeutic, it is shown herein that elimination of L-asparaginase activity in the immunostimulatory bacteria provided herein, and in other immunostimulatory bacteria and bacterial vaccines, such as the BCG vaccine used for vaccinating against tuberculosis (TB) and other diseases, enhances the function of T-cells in the tumor microenvironment. Elimination of L-asparaginase activity can be effected by modifying the bacterial genome to eliminate expression of active enzyme. Modifications include insertions, deletions, replacements, and other changes in the nucleic acids, so that the resulting encoded enzyme is not active, or not expressed, or is eliminated. It is shown herein that deletion of all or of a part of the gene that encodes L-asparaginase II, ansB, or disruption thereof, to eliminate expression of the encoded enzyme in the immunostimulatory bacteria, enhances the function of T-cells in a bacterially- colonized tumor microenvironment. Inhibition of L-asparaginase II activity is accomplished by deletion of all or of a part of, or interruption/disruption of, the gene ansB in the immunostimulatory bacteria, whereby L-asparaginase II is not produced. Thus, provided are immunostimulatory bacteria whose genomes are modified so that L-asparaginase II is not produced. Immunostimulatory bacteria provided herein are employed to colonize tumor-resident immune cells to enhance the anti-tumor immune response; included among the bacterial genome modifications are deletions, insertions, disruptions, and/or other modifications that eliminate the expression of L- asparaginase II.

As shown herein, the genome of the immunostimulatory bacteria herein can be modified to delete ansB, or to disrupt it or to otherwise modify it, to result in inactive encoded L-asparaginase II, or to eliminate the asparaginase, thereby preventing T-cell suppression, and enhancing the anti-tumor T-cell function in vivo. It is shown herein that strains in which ansB is intact induce profound T-cell immunosuppression in T- cells co-localized with the strain. Strains in which ansB is deleted do not induce immunosuppression, thus, solving another problem in the art in using bacteria to deliver encoded therapeutic products to tumors. Immunostimulatory bacteria that combine modifications, such as deletions or disruptions, of the ansB gene, whereby functional encoded enzyme is not expressed, with other modifications described herein that result in increased accumulation in the tumor microenvironment and/or in tumor-resident immune cells, provide superior therapeutic immunostimulatory bacteria.

Corresponding genes, encoding homologs or orthologs of L-asparaginase II (AnsB) in other bacterial species, also can be deleted or disrupted to achieve similar results. These genes include, but are not limited to, for example, ansB, encoding L- asparaginase 2 in A. coir, ansB (STY3259), encoding L-asparaginase in S. typhi,' ansB (lmol663), encoding asparagine synthetase in A. monocytogenes,' and BL1142, encoding an L-asparaginase precursor in Bifidobacterium longum.

6. Deletions/Disruptions in Salmonella Genes Required to Produce Curli Fimbriae

Bacteria and fungi are capable of forming multicellular structures called biofilms. Bacterial biofilms are encased within a mixture of secreted and cell wall- associated polysaccharides, glycoproteins, and glycolipids, as well as extracellular DNA, known collectively as extracellular polymeric substances. These extracellular polymeric substances protect the bacteria from multiple insults, such as cleaning agents, antibiotics, and antimicrobial peptides. Bacterial biofilms allow for colonization of surfaces, and are a cause of significant infection of prosthetics, such as injection ports and catheters. Biofilms also can form in tissues during the course of an infection, which leads to increases in the duration of bacterial persistence and shedding, and limits the effectiveness of antibiotic therapies. Chronic persistence of bacteria in biofilms is associated with increased tumorigenesis, for example in S. typhi infection of the gall bladder (see, e.g., Di Domenico et al. (2017) Int. J. Mol. Sci. 18:1887). csgD Deletion

In Salmonella, such as S. typhimurium, biofilm formation is regulated by the csgD gene, which activates the csgBAC operon, and results in increased production of the curli fimbriae subunits CsgA and CsgB (see, e.g., Zakikhany et al. (2010) Molecular Microbiology 77(3):771-786). CsgA is recognized as a PAMP by TLR2, and induces production of IL-8 from human macrophages (see, e.g., Tukel et al. (2005) Molecular Microbiology 58(l):289-304). Also, csgD indirectly increases cellulose production by activating the adrA gene that encodes for di-guanylate cyclase. The small molecule cyclic di-guanosine monophosphate (c-di-GMP), generated by adrA, is a ubiquitous secondary messenger that occurs in almost all bacterial species. Increases in c-di-GMP enhance expression of the cellulose synthase gene bcsA, which in turn increases cellulose production via stimulation of the bcsABZC and bcsEFG operons, leading to cellulose biofilm formation. As a result, bacteria, such as S. typhimurium, can form biofilms in solid tumors as protection against phagocytosis by host immune cells. Bacterial mutants, such as Salmonella mutants, that cannot form biofilms, are taken up more rapidly by host phagocytic cells, and are more readily cleared from infected tumors (see, e.g., Crull et al. (2011) Cellular Microbiology 13(8): 1223-1233). This increase in intracellular localization within phagocytic cells can reduce the persistence of extracellular bacteria, and, as shown herein, can enhance the effectiveness of plasmid delivery of therapeutic products, such as immunomodulatory proteins and other anti-cancer therapeutics, as described herein. Reduction in the capability of immunostimulatory bacteria, such as S. typhimurium, to form biofilms, can be achieved through deletion or disruption of genes involved in biofilm formation, such as, for example, csgD, csgA, csgB, adrA, bcsA, bcsB, bcsZ, bcsE, bcsF, bcsG, dsbA, or dsbB (see, e.g., Anwar et al. (2014) PLoS ONE 9(8):el06095).

It is shown herein that engineering the immunostimulatory bacteria to reduce biofilm formation increases clearance rates from tumors/tissues, increasing tolerability of the therapy, and prevents colonization of prosthetics in patients, thereby increasing the therapeutic benefit of these strains. It is known that adenosine mimetics inhibit S. typhimurium biofilm formation, indicating that the high adenosine concentration in the tumor microenvironment can contribute to tumor-associated biofilm formation (see, e.g., Koopman et al. (2015 ) Antimicrob. Agents Chemother. 59:76-84). It is shown herein that csgD-deleted immunostimulatory bacterial strains demonstrate improved anti-tumor efficacy because of greater bacterial uptake into tumor-resident myeloid cells. Similar genome modifications can be effected in other bacterial strains, such as E. coll and Listeria, that alter biofilm formation and/or curb fimbriae production, so that the bacteria can better infiltrate the vasculature, such as that present in tumors and the tumor microenvironment.

Corresponding genes, encoding homologs and orthologs of csgD, and other genes that are required for curli fimbriae and biofilm formation in other bacterial species, also can be deleted or disrupted or otherwise modified to achieve similar results. These genes include, but are not limited to, for example, csgD, encoding DNA-binding transcriptional dual regulator CsgD in . co!i, csgD (STY1179), encoding regulatory protein CsgD in S. typhr, and lep, encoding the Listeria cellulose binding protein that is involved in biofilm formation in L. monocytogenes.

Modification of the bacterial genome, such as by deletion or disruption of genes to render the bacteria csgD', results changes, such as one or more of elimination of curli fimbriae and inflammatory cyclic dinucleotides (CDNs), and removes cellulose secretion. This eliminates inflammatory and immunosuppressive elements, prevents TLR4 recognition through altered lipid A acylation, and eliminates cellulose secretion, and, thus, possible biofilm formation, thereby increasing safety and efficacy.

As described herein, bacterial strains, such as 5. typhimurium strains, that are engineered to be auxotrophic for adenosine; and are reduced in their ability to induce pro-inflammatory cytokines by modification of the LPS and/or deletion of flagellin; and/or that do not express L-asparaginase II to improve T-cell function; and/or that contain deletions or disruptions of genes required for biofilm formation; and/or that are further modified to maintain significant plasmid copy number per cell, at least low to medium copy number or higher, in the absence of antibiotic selection; and that deliver genetic expression cassettes encoding therapeutic products, promote robust anti-tumor immune responses. The plasmids include regulatory sequences to promote secretion of the encoded therapeutic products into the tumor microenvironment.

7. Improving Resistance to Complement

The complement system is the first line of immune defense against invading pathogens that directly activate the lectin pathway or the alternative pathway (AP) cascades in the human host. The complement system involves more than 30 soluble and cell-membrane bound proteins that function in the innate immune response to recognize and kill pathogens, such as bacteria, virus-infected cells, and parasites, and also, that play a role in the antibody-mediated immune response. Activation of the complement cascade leads to opsonization of foreign microbes, release of chemotactic peptides, and finally, to disruption of bacterial cell membranes. Three homologous glycoproteins in the complement system, C3, C4, and C5, play a central role in complement function and interact with other complement components. C3b and C4b, generated from C3 and C4, respectively, are important components of convertases that promote activation of the complement cascade. The cleavage fragments of C5 are C5a, which induces migration of phagocytes into the infection site, and C5b, which initiates the formation of the membrane attack complex (MAC), and bacterial lysis (see, e.g., Ramu et al. (2007) FEBS Letters 581:1716-1720).

To survive, pathogens have developed strategies to prevent deleterious consequences of complement activation. For example, members of the Ail/Lom family of outer membrane proteins provide protection from complement-dependent killing for a number of pathogenic bacteria. Members of the Ail/Lom family, which include Ail (attachment invasion locus) of Yersinia species, e.g., Y. enterocolitica and Y. pseudotuberculosis, Rck (resistance to complement killing) and PagC of Salmonella species, and OmpX of Escherichia coli, are outer membrane proteins that share significant amino acid sequence similarity and identity, and have similar membrane topologies. While members of this family of proteins exhibit diverse functions, several of them, including Ail of Y. enterocolitica and Y. pseudotuberculosis, as well as Rck of S. enterica, function, at least in part, to protect bacteria from complement-mediated lysis (see, e.g. , Bartra et al. (2008) Infection and Immunity 76:612-622).

Another bacterial product that aids in avoiding or mitigating complement is the surface protease, designated PgtE (outer membrane serine protease) in Salmonella, and other members of the omptin family. The surface protease PgtE of . enterica belongs to the omptin family of enterobacterial outer membrane aspartate proteases. PgtE and other omptins require rough LPS to be active, but are sterically inhibited by the O-antigen. Expression of pgtE is upregulated during the growth of Salmonella inside macrophages, and the bacteria released from macrophages exhibit strong PgtE- mediated proteolytic activity. PgtE proteolytically activates the mammalian plasma proenzyme plasminogen to plasmin, inactivates the mam physiological inhibitor of plasmin, alpha 2-antiplasmin, and mediates bacterial adhesion to extracellular matrices of human cells. This way, PgtE mediates the degradation of extracellular matrix components and generates potent, localized proteolytic activity, which can promote migration of Salmonella across extracellular matrices. PgtE also degrades alpha-helical antimicrobial peptides which can be important during intracellular growth of Salmonella. The omptin Pla of Yersinia pestis is a close ortholog of PgtE and shares functions with PgtE. Pla cleaves C3, and PgtE increases serum resistance of Salmonella by cleaving complement components C3b, C4b, and C5. The gene pgtE, and orthologs thereof from other bacterial species, can be included in the immunostimulatory bacteria herein to increase resistance to complement.

It is shown herein that the effects of complement in human serum explain the failure of therapeutic immunostimulatory bacteria, such as the Salmonella strain VNP20009, which had been shown to effectively colonize tumors in rodent models. Systemic administration of VNP20009 resulted in colonization of mouse tumors (see, e.g., Clairmont et al. (2000) ; Infect. Dis. 181: 1996-2002; and Bermudes etal. (2001) Biotechnol. Genet. Eng. Rev. 18:219-33); whereas systemic administration of VNP20009 in human patients resulted in very little colonization. In the Phase 1 Study in advanced melanoma patients, very little VNP20009 was detected in human tumors after a 30 minute intravenous infusion (see, Toso et al. (2002) J. Clin. Oncol. 20: 142- 52). Patients that entered into a follow-up study evaluating a longer, four-hour infusion of VNP20009, also demonstrated a lack of detectable VNP20009 after tumor biopsy (see, Heimann etal. (2003) J. Immunother. 26:179-180). Following intratumoral administration, colonization of a derivative of VNP20009 was detected (see, Nemunaitis etal. (2003) Cancer Gene Ther. 10:737-744). Direct intratumoral administration of VNP20009 to human tumors resulted in much higher tumor colonization, indicating that human tumors can be colonized at a high level, and that the difference in tumor colonization between mice and humans occurs only after systemic administration.

It is shown and described herein, that, while not previously known to occur in wild-type S. typhimurhim, VNP20009 is inactivated by human complement, which explains the low tumor colonization observed in humans upon systemic administration of VNP20009. Strains provided herein exhibit resistance to complement. They can be modified to express Rck and other proteins involved in mediating complement resistance or avoidance, such as Ail of Yersinia enterocolitica, or PgtE of Salmonella typhimurium, or, if they natively express such a protein, they can be modified to overexpress Rck and/or other such proteins. Rck can be introduced into bacteria, such as E. coli, that lack a homolog.

Rck Expression

Rck (resistance to complement killing) is a 17 kDa outer membrane protein encoded by the large virulence plasmid of Salmonella species, such as . enteritidis and S. typhimurium, that induces adhesion to and invasion of epithelial cells. The Rck protein has been shown to protect S. enterica from complement by inhibiting C9 polymerization and subsequent assembly of a functional membrane attack complex. An rck mutant exhibited a 2-3 fold decrease in epithelial cell invasion compared to the wild-type strain, while rck overexpression in wild-type strains leads to increased invasion. The Rck protein induces cell entry by a receptor-mediated process, promoting local actin remodeling, and weak and closely adherent membrane extensions. Thus, Salmonella can enter cells by two distinct mechanisms: the Trigger mechanism mediated by the T3SS-1 complex, and a Zipper mechanism induced by rck (see, e.g., Manon et al. (2012), Salmonella, Chapter 17, eds. Annous and Gurtler, Rijeka, pp. 339-364). Expression of rck on the Salmonella virulence plasmid confers a high level of resistance to neutralization by human complement, by preventing the formation of the membrane attack complex. When the A typhimurium virulence plasmid containing rck was expressed in a highly serum-sensitive strain of E. coli, Rck was able to restore complement resistance.

The immunostimulatory bacteria provided herein retain, or are provided with, Rck to confer resistance to human complement. It is shown herein that immunostimulatory bacteria, such as E. coli, can be modified by encoding rck on a plasmid in the bacteria to thereby confer resistance to complement. Immunostimulatory bacteria provided herein encode rck, either endogenously, or can be modified to encode it in order to increase resistance to complement. Methods for conferring resistance to complement also are provided. For example, the therapeutic E. coli species described in U.S. Patent Application Publication Nos. 2018/0325963 and 2018/0273956, and U.S. Patent Nos. 9,889,164 and 9,688,967, can be improved by modifying the bacteria therein, such as by introducing nucleic acid encoding the Salmonella rck gene on a plasmid therein, to thereby improve or provide resistance to complement. Bacteria that are resistant to complement can be systemically administered, and sufficient bacteria can survive to be therapeutically effective.

Nucleic acids encoding the Salmonella rck gene are introduced into bacteria, such as therapeutic E. coll, to thereby confer or increase complement resistance.

Other orthologs and homologs of rck from other bacterial species, similarly can be expressed in the immunostimulatory bacteria. For example, Ail is an Rck homolog from Yersinia enterocolitica, which enhances complement resistance under heterologous expression. PgtE is an S. typhimurium surface protease that has also been shown to enhance complement resistance under heterologous expression.

8. Deletions of Genes Required for Lipoprotein Expression in Salmonella and Other Gram-Negative Bacteria

The LPS and Braun (murein) lipoprotein (Lpp) are major components of the outer membrane of Gram-negative enteric bacteria that function as potent stimulators of inflammatory and immune responses. Braun (murein) lipoprotein (Lpp) is one of the most abundant components of the outer membrane in S. typhimurium, and leads to TLR2 induction of pro-inflammatory cytokines, such as TNFa, IL-6, and IL-8 (in humans). Two functional copies of the lipoprotein gene (IppA (SEQ ID NO:387) and IppB (SEQ ID NO:388), that are located on the bacterial chromosome of Salmonella, contribute to bacterial virulence. Deletion of the IppA and IppB genes, and elimination of lipoprotein expression, reduces virulence and decreases pro-inflammatory cytokine production (see, e.g., Sha et al. (2004) Infect. Immun. 72(7):3987-4003; and Fadi et al. (2005) Infect. Immun. 73(2): 1081-1096). Deletion of the lpp genes would be expected to reduce infection of cells, and, thus, decrease plasmid delivery and expression of the encoded therapeutic products or proteins. As shown in Example 18 below, however, while deletion of these genes did reduce tumor colonization, the amount of plasmid delivered to the targeted cells, i.e., the tumor-resident immune cells, particularly macrophages, significantly was increased. As shown herein, deletion or disruption of these genes (IppA and IppB), thus, results in decreased virulence due to the inability to survive in infected macrophages, but results in enhanced plasmid delivery by the immunostimulatory bacteria, thereby increasing the expression of encoded therapeutic genes in the targeted cells, i.e., the tumor-resident immune cells, particularly macrophages. Strains provided herein are AFLG (lack flagella), and/or ApagP. and/or Aans j, and/or AcsgD. Additionally, the strains are one or more of purl (JspurM), SmsbB, and Sasd (in the bacterial genome). Exemplary strains are , pnrl ypurM), EpagP, SansB, Sasd, or EthyA in place of dasd. The strains also can be lppA~ and/or IppB , particularly lppA~/lppB~. The plasmid is modified to encode therapeutic products under control of host-recognized promoters (e.g. , eukaryotic promoters, such as RNA polymerase II promoters, including those from eukaryotes, and from animal viruses). The plasmids can encode asd to permit bacterial replication in vivo, and can encode nucleic acids with other beneficial functions (such as CpGs), and can encode gene products, as described elsewhere herein.

The immunostimulatory bacteria provided herein can be modified to eliminate the ability to infect epithelial cells, such as by elimination of the flagella. Elimination of the ability to infect epithelial cells, as described elsewhere herein, also can be achieved by inactivating SPI-1 -dependent invasion, through inactivation or knockout of one or more genes involved in the SPI-1 pathway. These genes include, but are not limited to, one more of: avrA, hilA, hilD, invA, invB, invC, invE, invF, invG, invH, invl, invJ, iacP, iagB, spaO, spaP, spaQ, spaR, spaS, orgA, orgB, orgC,prgH,prgI, prgj, prgK, sic A, sicP, sip A, sipB, sipC, sipD, sirC, sopB, sopD, sopE, sopE2, sprB, and sptP. Additionally or alternatively, the immunostimulatory bacteria can contain knockouts or deletions in genes to inactivate products involved in SPI- 1 -independent infection/invasion, such as one or more of the genes fljB,fliC, rck,pagN, hlyE,pefl, srgD, srgA, srgB, and srgC, and/or the immunostimulatory bacteria can contain knockouts or deletions to inactivate products of genes that induce cell death of tumorresident immune cells, such as genes that encode proteins that are directly recognized by the inflammasome, including fljB,fliC,prgI (needle protein), and prgj (rod protein). The /c/r gene is desirable because it protects the bacteria against inactivation by complement. Bacteria that do not endogenously encode rck, can be modified to encode a heterologous rck gene.

The immunostimulatory bacteria are derived from suitable bacterial strains. Bacterial strains can be attenuated strains, or strains that are attenuated by standard methods, or that, by virtue of the modifications provided herein, are attenuated in that their ability to colonize is limited primarily to immunoprivileged tissues and organs, particularly tumor-resident immune cells, the TME, and tumor cells, including solid tumors. Bacteria include, but are not limited to, for example, strains of Salmonella, Shigella, Listeria, E. coli, and Bifidobacteriae . For example, species include Shigella sonnei, Shigella flexneri, Shigella dysenteriae, Listeria monocytogenes, Salmonella typhi, Salmonella typhimurium, Salmonella gallinarum, and Salmonella enteritidis. Other suitable bacterial species include Rickettsia, Klebsiella, Bordetella, Neisseria, Aeromonas, Francisella, Corynebacterium, Citrobacter, Chlamydia, Haemophilus, Brucella, Mycobacterium, Mycoplasma, Legionella, Rhodococcus, Pseudomonas, Helicobacter, Vibrio, Bacillus, and Erysipelothrix . For example, Rickettsia rickettsii, Rickettsia prowazekii, Rickettsia tsutsugamuchi, Rickettsia mooseri, Rickettsia sibirica, Bordetella bronchiseptica, Neisseria meningitidis, Neisseria gonorrhoeae, Aeromonas eucrenophila, Aeromonas salmonicida, Francisella tularensis, Corynebacterium pseudotuberculosis, Citrobacter freundii, Chlamydia pneumoniae, Haemophilus somnus, Brucella abortus, Mycobacterium intracellulare, Legionella pneumophila, Rhodococcus equi, Pseudomonas aeruginosa, Helicobacter mustelae, Vibrio cholerae, Bacillus subtilis, Erysipelothrix rhusiopathiae , Yersinia enter ocolitica, Rochalimaea quintana, and Agrobacterium tumerfacium.

Exemplary of the immunostimulatory bacteria provided herein are species of Salmonella. Exemplary of bacteria for modification as described herein are wild-type strains of Salmonella, such as the strain that has all of the identifying characteristics of the strain deposited in the American Type Culture Collection (ATCC) as accession #14028. Engineered strains of Salmonella typhimurium, such as strain YS1646 (ATCC catalog # 202165, also referred to as VNP20009; see, also International PCT Application Publication No. WO 99/13053), is engineered with plasmids to complement an asd gene knockout and to allow for antibiotic-free plasmid maintenance. The strains then are modified to delete the flagellin genes, and/or to delete pagP. The combination of flagella knockout and pagP deletion renders the strains highly resistant to human serum complement. The strains also are rendered auxotrophic for purines, particularly adenosine, and are asd~ and msbB'. As exemplified, strains in which the purl and msbB genes are completely deleted are more fit (e.g., grow faster) that strain VNP20009, in which these genes are not deleted, but are modified to eliminate expression. The asd gene can be provided on a plasmid for in vivo replication in the eukaryotic host. The strains also have a modification, such as a deletion, disruption, or other modification, in the ansB gene, preventing them from producing immunosuppressive L-asparaginase II, and improving tumor T-cell function. The strains also are modified to eliminate biofilm production, such as by a csgD deletion, which renders them unable to produce curb fimbriae, cellulose, and c-di-GMP, reducing unwanted inflammatory responses, and preventing the strains from forming biofilms.

These genomic deletions and plasmids are described and exemplified elsewhere herein. Any of the nucleic acid encoding therapeutic products, such as immunostimulatory proteins and other products, described elsewhere herein and/or known to those of skill in the art, can be included on the plasmid. The plasmid generally is present in low to medium copy number, as described elsewhere herein. Therapeutic products include gain-of-function mutants of cytosolic DNA/RNA sensors, that can constitutively evoke/induce type I IFN expression, and other immunostimulatory proteins, such as cytokines, chemokines, and co-stimulatory molecules, that promote an anti-tumor immune response in the tumor microenvironment, and other such products described herein. The plasmids also can encode antibodies, and fragments thereof, e.g., single-chain antibodies that target immune checkpoints and other cancer targets, such as VEGF, IL-6, and TGF-P, and other molecules, such as bispecific T-cell engagers, or BiTEs®. The plasmids also can encode IL-6 binding decoy receptors, TGF-beta binding decoy receptors, and TGF- beta polypeptide antagonists.

9. Robust Immunostimulatory Bacteria Whose Genomes are Optimized for Anti-Tumor Therapy, and that Encode Therapeutic Products, Including a Plurality Thereof

As described herein, bacterial strains, such as S. typhimurium strains, that are engineered to be adenosine auxotrophic, and are reduced in their ability to induce pro- inflammatory cytokines by modification of the LPS and/or deletion of flagellin, and/or are modified by deletion or elimination of L-asparaginase II expression to improve T-cell function, and/or are modified by deletion or disruption of genes required for biofilm formation, and/or that demonstrate enhanced human serum survival due to increased rck expression, are further modified to deliver therapeutic products, such as immunomodulatory proteins, and promote robust anti-tumor immune responses.

The table below summarizes the bacterial genotypes/modifications, with reference to Salmonella genes with the understanding that counterpart genes can be modified in other bacterial species, their functional effects, and some of the effects/benefits achieved herein.

As shown in the examples, immunostimulatory bacteria that comprise these genome modifications, such as the alterations to the bacterial surface coat, and others of these changes result in a reduction of TLR-mediated inflammatory cytokines, such as reduction in or elimination of induced levels of IL-10. It is shown herein that these bacteria are consumed by immune cells, and can deliver encoded payloads, such as immunostimulatory protein(s). These bacteria target (or a taken up by) phagocytic tumor-resident myeloid cells, such as CD45+ cells, tumor associated macrophages (TAMS), and dendritic cells.

10. Vaccines and Bacteria that Deliver RNA, including mRNA and other Forms of RNA, for Expression in a Eukaryotic Host

Immunostimulatory Bacteria as Delivery Vehicles for RNA

Provided are immunostimulatory bacteria that contain or that express encoded heterologous RNA, where the bacteria infect eukaryotic host cells, and release the encoded RNA into the cytoplasm of the host cell (see, e.g., U.S. Patent No. 7,390,646). The RNA can be a therapeutic product, or the RNA is translated by the host cell machinery into a therapeutic product. Such products include immunostimulatory proteins and antigens from pathogens and/or tumor antigens that can be expressed in cells, such as phagocytic cell for presentation to induce immune responses including T-cell responses, memory T-cells, and antibodies. For immunization, the bacteria can be administered to the site or by a route that is the same as naturally-occurring route of the pathogen, such as by inhalation or nasal administration for respiratory viruses, oral administration for gastric and enteric pathogens, intramuscular administration, such as for viruses, and also systemic routes. This can result in in situ immune responses that can provide lifelong immunity.

Any of the immunostimulatory bacteria provided herein can be modified so that the bacteria produce the exogenous or heterologous RNA within the bacterial cells in vitro, such as during in vitro cell culture, or after entry of the bacterial cells into the host eukaryotic cells, or both during in vitro cell culture, and after entry of the bacterial cells into the host eukaryotic cells. Following infection of eukaryotic host cells, the bacteria release the exogenous RNA into the cytoplasm of the host cells. The RNA can be encoded on a plasmid in the bacteria, and can encode a plurality of therapeutic products. In particular, the forms of RNA delivered are any that can be translated by eukaryotic cell machinery. This includes mRNA, long non-coding RNA, RNAi, dsRNA, circular RNA (eRNA; see, e.g., U.S. Patent No. 10,953,033).

Provided are immunostimulatory bacteria and genomically-modified bacteria that contain or express encoded heterologous RNA, where, when the bacteria infect eukaryotic host cells, the encoded RNA, such as mRNA, is released into the cytoplasm of the host cell. The bacteria include any of the immunostimulatory bacteria provided herein, and particularly, those that include genomic modifications, so that they have reduced TLR2, TLR4, and TLR5 signaling. The bacteria also can encode a therapeutic product or a plurality thereof, and particularly, encode immunostimulatory proteins, including cytokines and STING proteins, including the modified STING proteins provided herein, which can act as adjuvants if the RNA is encoding an antigen or protein for vaccination, or for a treatment for which immune stimulation is advantageous.

The nucleic acid encoding the RNA is modified so that it is transcribed in the bacteria, but is not translated. This can be effected by designing the nucleic acid so that the transcribed RNA product is not recognized by or cannot be translated by prokaryotic ribosomes. The encoding nucleic acid further is designed such that the transcript is recognized by and is translated by eukaryotic ribosomes, such as those present in a host, such as a human. This can be effected by, for example, including nucleic acid encoding an internal ribosome entry site (IRES) that is not recognized by prokaryotic ribosomes, but is recognized by eukaryotic ribosomes The encoded RNA typically is mRNA, but can be other forms of RNA that can be translated into therapeutic products or immunizing antigens or proteins. Other forms, of RNA include, but are not limited to, eRNA, which is circular RNA. eRNA is a circular polyribonucleotide that: (a) contains an expression sequence encoding a polypeptide, such as an antigen; (b) contains an internal ribosome entry site (IRES) that is not recognized by prokaryotic ribosomes, and a termination element; and (c) lacks a poly- A sequence, a free 3' end, and an RNA polymerase recognition motif (see, e.g., U.S. Patent No. 10,953,033, which describes such circular RNA). The bacteria are modified so that the transcribed RNA is circularized in vivo.

The RNA can be a therapeutic product, or the RNA is translated by the host cell machinery into a therapeutic product. The RNA can encode an antigen, whereby it is encoded into a protein when delivered into the host cells, to immunize the host against the antigen, and the pathogen or the tumor from which the antigen is derived. The nucleic acid can be modified so that the encoded antigen is modified to improve its properties for immunization, such as by stabilizing a conformation of the antigen against which antibodies and other adaptive immune responses are directed. This has been done, for example, for the spike proteins, on corona viruses, so that the conformation that binds to host cell receptors is stabilized, resulting in a more robust antibody response.

Any of the immunostimulatory bacteria provided herein can be modified so that the bacteria produce the exogenous or heterologous RNA within the bacterial cells in vitro, such as during in vitro cell culture, or after entry of the bacterial cells into the host eukaryotic cells, or both during in vitro cell culture and after entry of the bacterial cells into the host eukaryotic cells Following infection of eukaryotic host cells, the bacteria release the exogenous RNA into the cytoplasm of the host cells. The RNA can be encoded on a plasmid in the bacteria, and can encode a plurality of therapeutic products.

The nucleic acid encoding the RNA is operably linked to a promoter recognized by the bacteria, or the bacteria are modified to encode the RNA polymerase that recognizes the promoter, such as a bacterial promoter, or a bacteriophage promoter that is recognized by phage T7 RNA polymerase. The RNA can be mRNA, or other RNA molecules that contain regulatory sequences for expression in a eukaryotic host cell. The promoter can be an inducible promoter. The bacteria can contain a plurality of copies of the encoded RNA, which, when introduced into a eukaryotic host, such as a human, is released into the host cell cytoplasm, where, if it is mRNA, it can be translated. Bacterial transcription is decoupled from bacterial translation, so that gene products are transcribed, but not translated. In some embodiments, rather than including an IRES sequence, the nucleic acid encoding the RNA lacks a Shine-Dalgamo (SD) sequence upstream of the start codon, whereby the RNA cannot be translated by bacterial or archaeal ribosomes, but can be translated by eukaryotic ribosomes. The RNA can contain sequences, such as a Kozak sequence, with a start codon recognized by eukaryotic ribosomes. Shine- Dalgamo sequences include, for example aggagg (SEQ ID NO:389), agga (SEQ ID NO:390), gagg (SEQ ID NO:391), and ggag (SEQ ID NO:392):

‘’‘..AUAAAGGAGGUAAAUA A; JG -

...AUAAAGGAAAUAAAUAAiJG - *•

...AUAACAGAGGUAAAUAAGG - w

...AUAACAGGAGUAAAUAAUG •"

(see, e.g., parts. igem.org/File:RBSAlignedSpacing. png). The RNA also can contain one or more Internal Ribosome Entry Site(s) (IRES), which block or inhibit translation by prokaryotic ribosomes. Thus, the RNA is produced in the bacteria, which infect or have infected myeloid cells, thereby delivering RNA to the myeloid cells, where, if it is mRNA, it is translated in the myeloid cells, which recognize the RNA for translation because of the Shine-Dalgamo sequence, or IRES, or other such regulatory sequence. Inserting, for example, an IRES between a promoter and a gene sequence, de-couples transcription from translation The immunostimulatory bacteria, thus, serve as an RNA delivery system.

In embodiments, bacterial systems, such as the immunostimulatory bacteria provided herein, are used to deliver genetic payloads to tissue-resident and/or tumorresident myeloid cells. In this context, the bacteria are dosed in vivo, the bacteria are, for example, asd , and do not include a plasmid containing a complementing asd gene cassette, so that they cannot replicate in vivo. To grow the bacteria, they are cultured in vitro, diaminopimelic acid (DAP) is added to the culture medium to facilitate bacterial replication in the absence of a functional asd gene. In this context, when such immunostimulatory bacteria are administered, one bolus of nucleic acid is delivered to tissue-resident or tumor resident myeloid cells. After phagocytosis and intracellular destruction of the bacteria, the RNA encoding the therapeutic product(s) is released into the cytoplasm of the myeloid cells for translation, to produce the therapeutic product(s) that is/are secreted from the myeloid cells into the surrounding environment, such as the tumor microenvironment, when the infected cells are tumorresident myeloid cells.

Bacteria cultured in vitro encode RNA, but lack the signal s/regulatory sequences, such as the Shine-Dalgamo sequence, that are required to translate the RNA, or the encoding nucleic acid contains sequences, such as an IRES, that blocks or does not provide for translation by prokaryotic ribosomes, but that are recognized by eukaryotic ribosomes, so that the bacteria produce, but do not translate, the mRNA. The bacteria, which are modified as described herein, so that they infect or accumulate in tissue-resident or tumor-resident immune cells, particularly tissueresident or tumor-resident myeloid cells, deliver the RNA to the myeloid cells, where it is translated. Any RNA can be delivered, including RNA encoding antigens, such as from pathogens or tumors, and/or encoding anti-viral therapeutics, which are expressed under control of a prokaryotic promoter. For such applications, the plasmids can be present in higher copy number (e.g., 150 copies or greater, such as 200, 300, 400, 500, or more copies, such as 500-700 copies), so that large amounts of the RNA is produced and delivered to the myeloid cells.

Also, as described and exemplified herein and discussed above, the encoding and regulatory sequences can be prepared so that the RNA is produced in the bacterium, but the encoding nucleic acid includes regulatory sequences that inhibit or prevent translation by bacterial ribosomes, but that are recognized by, and/or enhance, promote, or permit translation by, eukaryotic host ribosomes. For example, the mRNA is transcribed under control of a prokaryotic promoter. An IRES sequence is included in the transcript, which blocks translation in the prokaryotic bacterium, but, when delivered into a eukaryotic host cell, permits translation by eukaryotic ribosomes. The bacteria, thus, provide non-labile mRNA delivery systems. The bacteria can be grown in large quantities, and then can be lyophilized or frozen and/or stored at room temperature. They can be formulated into tablets, or powders, or can be micronized for inhalation.

The bacteria include those that contain genome modifications, whereby the response by toll-like receptors (TLRs) 2, 4, and 5 is reduced, compared to the bacterium without the genome modifications. These bacteria optionally can additionally contain genomic modifications so that they are auxotrophic for a required nutrient or factor, so that they are unable to replicate in a eukaryotic host, but can replicate in vitro when supplied with the nutrient or factor. These include deficiencies in adenosine biosynthesis, which has the added advantage of reducing immunosuppression when used for treatment of subjects with cancer who have tumors and a tumor microenvironment in which adenosine accumulates. Other deficiencies include thymine synthesis deficiencies, such as thyA described and discussed in sections above, and exemplified below.

The bacteria contain a plasmid containing nucleic acid encoding a product, or comprise RNA encoding the product. The plasmid can be present in low, medium, or high copy number. When used to deliver RNA, the plasmid can be present in high copy number, such as 150 copies or greater. The product encoded by the nucleic acid or RNA is an antigenic sequence or sequences from a pathogenic virus, bacterium, parasite, or is a tumor antigen, whereby, upon expression of the encoded antigen in the host, the host develops an immune-protective response or immunizing response against the pathogenic virus, bacterium, or parasite, or against the tumor antigen, or the product is a therapeutic product to be delivered to the host. Expression of the antigenic sequence(s) is/are under control of a prokaryotic promoter, so that RNA encoding the antigen(s) is produced in the bacterium, and, for RNA delivery, nucleic acid encoding the antigen comprises regulatory sequences that inhibit or prevent translation of encoded RNA by bacterial ribosomes, but that do not inhibit or prevent translation of the encoded RNA by eukaryotic host ribosomes, whereby translation is de-coupled from transcription in the bacterium. By virtue of the genomic modifications, the resulting bacterium is selective for infecting phagocytic cells when administered to a eukaryotic subject, and delivers the nucleic acid into the phagocytic cells, wherein the RNA is translated.

Of particular interest are the immunostimulatory bacteria provided herein that are modified so that they infect phagocytic cells, such as macrophages. In particular, the bacteria, which can be any suitable species/genus, such as Salmonella, Listeria, and E. coli, are so-modified so that they are one or more or all of modified to produce LPS with penta-acylated lipid A, modified to lack flagella, and modified to lack curli fimbriae; for example, the bacteria are msbRIpagP', flagellin", and csgD' (by modification by each of the encoding genes or equivalent genes in a particular species, so that the products are not produced, or, if produced, are inactive). The bacteria also are modified, such as by rendering them asd, so that the can be grown in vitro by providing the appropriate nutrient, such a DAP, but so that they do not replicate in a eukaryotic host, and ultimately die when the mRNA is delivered into the eukaryotic cells. The bacteria also can encode additional payloads, as described herein, such as modified STING proteins and/or cytokines (all as described and listed elsewhere herein), which serve to stimulate the host immune system, thereby acting as adjuvants. Such products can be encoded on plasmids in the bacterium, and can be encoded on a polycistronic construct with the mRNA encoding the antigen, and/or can be encoded under control of a eukaryotic promoter for transcription/translation into proteins, as described throughout the disclosure herein. These proteins can be designed for transcription, or, as appropriate, can be modified so that they are membrane anchored for presentations. The immunostimulatory bacteria, such as E. coli, can be modified so that they encoded and express rck to increase resistance to complement.

11. Bacterial Vaccines Against Particular Antigens, Including from Pathogens, and also Antigens from Tumors, for use as Anti-Pathogen Treatments and Vaccines, and For Anti-Cancer Treatment and/or Prevention

Vaccines can be used for preventing (reducing the risk) of contracting a disease or developing a disorder or cancer or as treatments for diseases, disorders, or cancers, and combinations thereof. Immunostimulatory bacteria provided herein and other vaccines can be used for these purposes. Because of the similarity in the immune response between an anti-tumor response and an anti-viral response, the immunostimulatory bacteria provided herein also can be used to treat infectious diseases. The bacteria can encode an anti-viral or anti-bacterial therapeutic, such as an inhibitor of a viral or bacterial product, or an inhibitor of the expression of a viral or bacterial product, or a viral or bacterial antigen. The combination of the immune response from the immunostimulatory bacteria and the therapeutic anti-pathogen product, and also, to the immunostimulatory proteins and encoded immune-stimulating proteins, provides a therapeutic immunostimulatory bacterium for vaccinating against and/or treating infectious diseases, particularly diseases associated with viral infections, such as chronic viral infections and latent viral infections. Of interest are chronic viral infections, such as infections by hepatitis viruses, herpesviruses, varicella zoster virus (VZV), Epstein-Barr virus, human immunodeficiency virus (HIV), human T-cell leukemia virus (HTLV), Respiratory Syncytial Virus (RSV), measles virus, and other such viruses that chronically infect subjects. The immunostimulatory bacteria also can be used for the treatment of acute infections as well, such as initial infections with chronic influenza and coronaviruses, such as Severe Acute Respiratory Syndrome coronavirus (SARS-CoV), Middle East Respiratory Syndrome coronavirus (MERS- CoV), and Severe Acute Respiratory Syndrome coronavirus 2 (SARS-CoV-2, which causes COVID-19). The immunostimulatory bacteria can encode an antigen from a pathogen, such as a viral antigen, and are used as a vaccine to prevent infection or to treat existing infections. The immunostimulatory bacterium, by virtue of its ability to accumulate in immune cells, such as antigen-presenting cells, such as T-cells, can promote a T-cell response to the virus. For example, immunostimulatory bacteria provided herein are deficient in asparaginase II, which enhances the function of T- cells to thereby promote the anti-viral response. The combination of expression of a viral antigen, such as an antigen from an essential viral protein, such as, for example, in the case of a coronavirus, an antigen from the Nucleocapsid, M and/or S proteins, which can result in neutralizing antibodies, and the enhancement of long-lived circulating and tissue-resident CD8 ⁺ T-cells, and provides effective vaccines and treatments.

The efficacy of existing vaccines, such as inactivated or attenuated vaccines, such as the Bacillus Calmette-Guerin (BCG) vaccine for tuberculosis (TB), can be improved by modifying the genome so that asparaginase is not expressed, thereby eliminating the immune suppression. BCG vaccine effectiveness is often reduced. This is by virtue of the asparaginase secreted by the vaccine bacteria, which decreases T-cell activity, thereby having an immuno-suppressive effect. This can be mitigated by modifying the bacteria to reduce or eliminate asparaginase activity. As discussed elsewhere herein, other modifications that reduce or eliminate TLR2 or TLR2/4/5 responses to the bacteria eliminate any blockade or inhibition of type I IFN by the activity or response of TLR2, or TLR4, or TLR5, or other toll-like receptors to bacteria. As described herein, the responses of the TLRs to bacteria can block or inhibit type I IFN; modifications of the bacteria herein by eliminating responses of these TLRs eliminate this undesirable effect.

As discussed in the section above, the immunostimulatory bacteria can encode an anti-viral therapeutic or an anti-bacterial therapeutic, or an antigen for immunization. Such therapeutics include inhibitors of viral genes and proteins, such as proteins required for replication and/or packaging, or the immunostimulatory bacteria can encode a therapeutic that prevents binding or interaction of a virus with a receptor or receptors that facilitate or provide for viral entry into a target cell. In some embodiments, the therapeutic protein can be under the control of a prokaryotic promoter. As discussed in the section above, the bacteria can be those that deliver RNA.

As discussed above, the bacteria can be used to deliver RNA, including mRNA for vaccination against pathogens, such as SARS-CoV, SARS-CoV-2, influenza, and other pathogens, and also, for anti-tumor therapy. The bacteria can also encode immunostimulants or other immune system enhancers, or inhibitors of immunosuppressors, such as immune checkpoint proteins. Exemplary of these are STING proteins, including the modified STING proteins provided herein, cytokines, and other such immune stimulating proteins. Any of the payloads described herein, including the combined payloads, that are for treating tumors, also can be encoded/ delivered by the bacteria. Anti-tumor responses and anti-viral responses are similar. For the vaccines, the immunostimulatory payloads serve as adjuvants to enhance the immune response. The pathogen antigen-encoding nucleic acid is under control of a prokaryotic promoter; the other payloads also can be provided as a polycistronic construct with the pathogen antigen-encoding nucleic acid, or they can be encoded under control of a eukaryotic promoter, as described herein for the antitumor therapeutics. Bacteria, such as those provided herein, that deliver payloads to phagocytic cells, and that contain the various genome modifications that improve their properties for delivery, can be used to deliver RNA The bacteria are superior vehicles to nanoparticles and other such delivery vehicles, since the RNA is delivered in the bacterial cytoplasm and contains the adjuvants to stimulate type I IFN. Among the advantages of the bacteria provided herein for delivery of RNA are: 1) they allow for RNA delivery “without the RNA” — the bacterial delivery vehicle obviates the need for complex RNA stability/manufacturing; 2) they result in direct targeting of phagocytic antigen presenting-cells (APCs); 3) they allow for the inclusion of payloads that provide type I IFN adjuvant activity; 4) they contain genome modification so that the bacteria do not replicate in the eukaryotic (such as human) host, such as by rendering the bacteria asd by inactivating endogenous asd activity and not encoding it on the plasmid; and 5) removal of bacterial products that are poor adjuvants by genome modification is possible, such as by eliminating the flagella, modifying the LPS, reducing or eliminating biofilm formation, as well as the elimination of asparaginase II activity. Exemplary of these bacteria are those that are msbBdpagP' , and flagellin", and optionally, ansB', and csgD', with a complete (clean) deletion of the purl gene. Bacteria as vehicles also permit rapid engineering and deployment; bacteria are easily scalable and are easy to manufacture; a plurality of different RNAs can be encoded in a single bacterium; and bacteria are stable, and they can be stored at room temperature, they can be lyophilized, and they can be formulated for intranasal delivery or delivery by inhalation, as well as for intravenous delivery, and can be formulated for other suitable routes of delivery. Retaining some of the TLR2 activity renders endothelial vascular cells leaky, as does including the genome modification that renders the bacteria csgD'.

For delivery of anti-cancer therapeutic products to the tumor/tumor microenvironment, the encoded products can be operatively linked to trafficking signals, such as signals for secretion. The products also can be designed for expression on a cell surface, such as the cell surfaces of tumor-resident myeloid cells and other phagocytic cells. The gene products can be modified so that they are membrane anchored. For example, IL- 12 has been modified herein so that it is membrane-anchored by adding a transmembrane domain to the C-terminus. Other proteins, including, for example, IL-2, IL-12, IL-12p35, IL-21, IL-15, and FLT-3L, can be similarly modified by adding a transmembrane domain or other such anchoring domain, such as a GPI anchor. Anchoring such proteins to the membranes of cells that are infected with the bacteria reduces toxicity so that they are not systemically secreted, but act in the tumor microenvironment. In particular, the immunostimulatory bacteria that encode such products and combinations thereof are those with genome modifications, such as, for example, modifications that result in the bacteria being msbB'/pagP' and not having flagella, whereby the response by toll-like receptors (TLRs) 2, 4, and 5 is reduced, compared to the bacteria without the genome modifications.

12. Conversion of M2 Phenotype Macrophages into Ml and Ml-Like Phenotype Macrophages

As described herein, the immunostimulatory bacteria provided herein accumulate in and/or target macrophages. Macrophages are phagocytic immune cells; they play a role in clearing senescent and apoptotic cells, as well as in the phagocytosis of immune-related complexes and pathogens, and in the maintenance of homeostasis. The phenotype and function of macrophages can be polarized by the microenvironment. There are two types: Ml -type (classically activated) macrophages, and M2 -type (alternatively activated) macrophages.

The role of Ml macrophages is to secrete pro-inflammatory cytokines and chemokines, and to present antigens, and thus, to participate in the positive immune response, and function as immune monitors. Ml macrophages produce pro- inflammatory cytokines, including IL-6, IL- 12, and TNF-alpha. M2 macrophages secrete arginase 1, IL-10, TGF-P, and other anti-inflammatory cytokines, which have the function of reducing inflammation, and contributing to tumor growth and immunosuppressive function.

As described elsewhere herein, conversion of the macrophages, whatever their initial phenotype, into a hybrid M1/M2 phenotype results in macrophages with an anti-tumor phenotype, and can convert cold (or immune desert) tumors into hot tumors that are susceptible to treatment, such as immunotherapy. Ml macrophage phenotypic markers include CD80 (also known as B7, B7.1, or BB1), CD86 (also known as B7.2), CD64 (also known as high affinity immunoglobulin gamma Fc receptor I ), CD16, and CD32 (also known as low affinity immunoglobulin gamma Fc region receptor lib). Expression of nitric oxide synthase (iNOS) in Ml macrophages can also serve as a phenotypic marker. CD 163 and CD206 are markers for the identification of M2 macrophages. Arginase 1 (Argl) and Dectin-1 also are phenotypic indicators for the identification of M2 macrophages. The phenotypic conversion can be monitored or assessed by virtue of expression of these markers, and/or other markers that are characteristic of such macrophage phenotypes.

Tumor-associated macrophages (TAMs) are associated with an immunosuppressive M2 phenotype. Immunostimulatory bacteria and other therapeutics provided herein can convert such macrophages into macrophages that have an anti-tumor phenotype. For example, immunostimulatory bacteria provided herein, that encode a therapeutic product that leads to expression of type I interferon (IFN), can effect such conversion. This property unique to the immunostimulatory bacteria, and other therapeutics by virtue of identification of the phenotype, provided herein, exploits the ability of the bacteria that include genomic modifications that result in the infection of macrophages. The encoded therapeutic products include those that are part of a cytosolic DNA/RNA sensor pathway, such as the STING variants (described in detail herein). The ability to convert macrophage phenotypes is demonstrated and exemplified in the Examples below. The expression of a modified STING protein by immunostimulatory bacteria provided herein that infect macrophages and express the STING protein, converts the phenotype of M2 macrophages into an anti-tumor phenotype.

Immunostimulatory bacteria provided herein, that include genome modifications as described herein, such as the elimination of flagella and LPS modification, convert infected M2 macrophages into those that induce some or all of the cytokine profiles that are anti-tumor. Immunostimulatory bacteria that express a variant STING protein that results in constitutive type I IFN expression in human primary M2 macrophages can convert them into macrophage that are susceptible to anti-cancer treatment, such as immunotherapy.

D. IMMUNOSTIMULATORY BACTERIA WITH ENHANCED THERAPEUTIC INDEX ENCODING GENETIC PAYLOADS THAT STIMULATE THE IMMUNE RESPONSE IN THE TUMOR MICROENVIRONMENT

The immunostimulatory bacteria provided herein are modified so that they accumulate in the tumor microenvironment, and in tumor-resident myeloid cells, where encoded therapeutic products, under the control of eukaryotic promoters, are expressed. The bacteria encode therapeutic products, particularly anti-cancer products, including products that stimulate the immune system, and/or that reverse or mitigate the immunosuppressive effects of tumors. As described herein, the bacteria can encode a plurality of products, where expression of each product is under the control of a separate promoter, or expression of the products is under the control of one promotor, and the expression cassette can include sequences that result in expression of the discrete products, and, where appropriate, includes regulatory sequences to ensure secretion of the encoded products into the tumor microenvironment. The immunostimulatory bacteria express encoded therapeutic products on the plasmid. As discussed herein, the plasmid can encode one product, or a plurality thereof. Each product can be expressed under the control of a different eukaryotic promoter, or multiple encoded products can be expressed under the control of a single promoter, such as by including 2A self-cleaving peptides between the coding portions, such as T2A (SEQ ID NO:327), P2A (SEQ ID NO:328), E2A (SEQ ID NO:329), and F2A (SEQ ID NO:330). The encoded products include those described herein, and they can be anti-cancer immune stimulating products whose activities are complementary. The immunostimulatory bacteria provided herein permit the combinatorial administration of multiple immunomodulatory products or payloads (multiplexed payloads) that would otherwise be too toxic if systemically administered. Exemplary of multiplexed payloads include one or more cytokine(s), an immunostimulatory protein to stimulate or induce expression of type I IFN, such as STING or a variant thereof that has increased activity or that is constitutively active, and a co-stimulatory molecule, such as an engineered 4-1BBL co- stimulatory molecule.

The immunostimulatory bacteria provided herein have strong anti-tumor effects, including provision of cures, such as after IV dosing with the multiplexed payloads or single agent payloads. The immunostimulatory bacteria, when systemically administered, infiltrate and enrich in solid tumors, in the TME, and in tumor-resident myeloid cells, in which the encoded therapeutic products are expressed and then locally delivered to the tumor microenvironment. Upon consumption (phagocytosis) by tumor-resident myeloid cells, the bacteria deliver a genetic payload-encoding plasmid, which allows for ectopic, single or multiplexed payload expression in a tumor-specific manner.

1. Immunostimulatory Proteins

The immunostimulatory bacteria herein can be modified to encode one or more of an immunostimulatory protein that promotes, induces, or enhances an antitumor response. As exemplified and described herein, the order in which the encoding nucleic acids are arranged on the plasmid can improve overall expression, and modifications to the plasmids, such as complete deletions or inactivation of genes, can improve the fitness of the bacteria that contain the plasmids encoding the proteins.

The immunostimulatory protein(s) can be encoded on a plasmid in the bacterium, under the control of a eukaryotic promoter, such as a promoter recognized by RNA polymerase II, for expression in a eukaryotic subject, particularly the subject for whom the immunostimulatory bacterium is to be administered, such as a human. The nucleic acid encoding the immunostimulatory protein(s) can include, in addition to the eukaryotic promoter, other regulatory signals for expression or trafficking in the cells, such as for secretion or expression on the surface of a cell.

Immunostimulatory proteins are those that, in the appropriate environment, such as a tumor microenvironment (TME), can promote, or participate in, or enhance, an anti-tumor response by the subject to whom the immunostimulatory bacterium is administered. Immunostimulatory proteins include, but are not limited to, cytokines, chemokines, and co-stimulatory molecules. These include cytokines, such as, but not limited to, IL-2, IL-7, IL-12, IL-15, IL-18, IL-21, IL-23, IL-12p70 (IL-12p40 + IL- 12p35), IL-15/IL-15R alpha chain complex, IL-36y, GM-CSF, IFNot, IFNp, IL-2 that has attenuated binding to IL-2Ra, and IL-2 that is modified so that it does not bind to IL-2Ra; chemokines, such as, but not limited to, CCL3, CCL4, CCL5, CXCL9, CXCL10, and CXCL11; and/or co-stimulatory molecules, such as, but not limited to, CD40, CD40L, 0X40, OX40L, 4-1BB, 4-1BBL, 4-1BBL with the cytoplasmic domain truncated or deleted (4-lBBLACyt), members of the TNF/TNFR superfamily (e.g., CD27 and CD27L), and members of the B7-CD28 family (e.g., CD80, CD86, ICOS, and ICOS ligand (B7RP1)).

Other such immunostimulatory proteins, that are used for the treatment of tumors, or that can promote, enhance or otherwise increase or evoke an anti -tumor response, that are known to those of skill in the art, are contemplated for encoding in the immunostimulatory bacteria provided herein. For example, the immunostimulatory bacteria can deliver a genetic payload encoding a truncated co- stimulatory molecule (e.g., 4-1BBL, CD80, CD86, CD27L, B7RP1, and OX40L), with a full or partial cytoplasmic domain deletion, for expression on an APC, where the truncated gene product is capable of constitutive immuno-stimulatory signaling to a T-cell through co-stimulatory receptor engagement, and is unable to counter- regulatory signal to the APC due to a deleted or truncated cytoplasmic domain. As described elsewhere herein, the modified truncated cytoplasmic domain, for example, of 4-1BBL, contains particular residues to ensure proper orientation of the protein domains, which increases expression of the protein. Deletion (full or partial), and modification, of the cytoplasmic domain of co-stimulatory molecules, potentiates the activation of the co-stimulatory molecule, without the immunosuppressive reverse signaling. a. Cytokines and Chemokines

In some embodiments, the immunostimulatory bacteria provided herein are engineered to express cytokines to stimulate the immune system, including, but not limited to, for example, IL-2, IL-7, IL-12, IL-12p70 (IL-12p40 + IL-12p35), IL-15, the IL-15/IL-15R alpha chain complex (IL-15Ra-IL-15sc), IL-18, IL-21, IL-23, IL- 36y, IL-2 that has attenuated binding to IL-2Ra, IL-2 that is modified so that it does not bind to IL-2Ra, IFN-a, and IFN- . Cytokines stimulate immune effector cells and stromal cells at the tumor site, and enhance tumor cell recognition by cytotoxic cells. In some embodiments, the immunostimulatory bacteria can be engineered to express chemokines, such as, for example, CCL3, CCL4, CCL5, CXCL9, CXCL10, and CXCL11.

IL-2

Interleukin-2 (IL-2), which was the first cytokine approved for the treatment of cancer, is implicated in the activation of the immune system by several mechanisms, including the activation and promotion of cytotoxic T lymphocyte (CTL) growth, the generation of lymphokine-activated killer (LAK) cells, the promotion of regulatory T-cell (Treg cell) growth and proliferation, the stimulation of tumor-infiltrating lymphocytes (TILs), and the promotion of T-cell, B cell, and NK cell proliferation and differentiation. Recombinant IL-2 (rIL-2) is FDA-approved for the treatment of metastatic renal cell carcinoma (RCC) and metastatic melanoma (see, e.g., Sheikhi etal. (2016) Iran J. Immunol. 13(3):148-166).

IL-7

IL-7, which is a member of the IL-2 superfamily, is implicated in the survival, proliferation, and homeostasis of T-cells. Mutations in the IL-7 receptor have been shown to result in the loss of T-cells, and the development of severe combined immunodeficiency disease (SCID), highlighting the critical role that IL-7 plays in T- cell development. IL-7 is a homeostatic cytokine that provides continuous signals to resting naive and memory T-cells, and which accumulates during conditions of lymphopenia, leading to an increase in both T-cell proliferation and T-cell repertoire diversity. In comparison to IL-2, IL-7 is selective for expanding CD8 ⁺ T-cells over CD4 ⁺ FOXP3 ⁺ regulatory T-cells. Recombinant IL-7 has been shown to augment antigen-specific T-cell responses following vaccination and adoptive cell therapy in mice. IL-7 also can play a role in promoting T-cell recovery following chemotherapy of hematopoietic stem cell transplantation. Early phase clinical trials on patients with advanced malignancy have shown that recombinant IL-7 is well-tolerated and has limited toxicity at biologically active doses (i.e., in which the numbers of circulating CD4 ⁺ and CD8 ⁺ T-cells is increased by 3-4 fold) (see, e.g., Lee, S. and Margolin, K. (2011) Cancers 3:3856-3893). IL-7 has been shown to possess antitumor effects in tumors, such as gliomas, melanomas, lymphomas, leukemia, prostate cancer, and glioblastomas, and the in vivo administration of IL-7 in murine models resulted in decreased cancer cell growth. IL-7 also has been shown to enhance the antitumor effects of IFN-y in rat glioma tumors, and to induce the production of IL-la., IL-ip, and TNF-a by monocytes, which results in the inhibition of melanoma growth. Additionally, administration of recombinant IL-7 following the treatment of pediatric sarcomas resulted in the promotion of immune recovery (see, e.g., Lin et al. (2017) Anticancer Research 37:963-968).

IL-12 (IL-12p70 (IL-12p40 + IL-12p35))

Bioactive IL-12 (IL-12p70), which promotes cell-mediated immunity, is a heterodimer, composed of p35 and p40 subunits, whereas IL-12p40 monomers and homodimers act as IL-12 antagonists. IL-12, which is secreted by antigen-presenting cells, promotes the secretion of IFN-y from NK and T-cells, inhibits tumor angiogenesis, results in the activation and proliferation of NK cells, CD8 ⁺ T-cells, and CD4 ⁺ T-cells, enhances the differentiation of naive CD4 ⁺ T-cells into Thl cells, and promotes antibody-dependent cell-mediated cytotoxicity (ADCC) against tumor cells. IL-12 has been shown to exhibit antitumor effects in murine models of melanoma, colon carcinoma, mammary carcinoma, and sarcoma (see, e.g., Kalinski et al. (2001) Blood 97:3466-3469; Sheikhi etal. (2016) Iran J. Immunol. 13(3): 148-166; and Lee, S. and Margolin, K. (2011) Cancers 3:3856-3893).

IL-15 and IL-15:IL-15Ra (IL-15/IL-15Ra)

IL-15 is structurally similar to IL-2, and while both IL-2 and IL-15 provide early stimulation for the proliferation and activation of T-cells, IL-15 blocks IL-2 induced apoptosis, which is a process that leads to the elimination of stimulated T- cells and the induction of T-cell tolerance, limiting memory T-cell responses, and potentially limiting the therapeutic efficacy of IL-2 alone. IL-15 also supports the persistence of memory CD8 ⁺ T-cells for maintaining long-term antitumor immunity, and has demonstrated significant antitumor activity in pre-clinical murine models via the direct activation of CD8 ⁺ effector T-cells in an antigen-independent manner. In addition to CD8 ⁺ T-cells, IL-15 is responsible for the development, proliferation, and activation of effector natural killer (NK) cells (see, e.g.. Lee, S. and Margolin, K. (2011) Cancers 3:3856-3893; and Han etal. (2011) Cytokine 56(3):804-810).

IL-15/IL-15R alpha chain complex can be used instead of IL- 15 because it has longer in vivo stability than monomeric IL-15, is 10- to 100-fold more active on immune cells than IL-15. IL-15/IL-15Ra stimulates the maturation, proliferation and anti-apoptotic maintenance of NK cells, \8 (gamma delta) T-cells and T-cells; and it promotes the survival of activated T-cells (CTLs) and generation of long-lived CD8+ CD44 ¹¹ ' T-cell memory cells. Clinically, IL-15 is superior to IL-2 as it does not bind IL-2Ralpha, located on immunosuppressive Tregs, and does not induce activation- induced cell death (AICD) or extensive capillary leak syndrome. It is shown herein that there is synergy between the type I interferon, induced by the engineered STING, to enhance CD8+ cell functions. There is synergy to enhance myeloid-mediated T-cell recruitment and CD8+ T-cell function. They act synergistically to activate human dendritic cells (DCs) to induce anti-viral CD8+ T-cell responses to thereby induce anti-cancer immunity. It is shown herein that this combination induces a high cure rate in an animal model. Strong T-cell infiltration in murine orthotopic T-cell excluded tumors, and breakdown of the stromal tumor associated macrophage (TAM) barrier. Significant T-cell and NK recruitment were observed.

IL- 15 and IL- 15 receptor alpha (IL-15Ra) are coordinately expressed by antigen-presenting cells, such as monocytes and dendritic cells, and IL- 15 is presented in trans by IL-15Ra to the TL-15Rpyc receptor complex expressed on the surfaces of CD8 ⁺ T-cells and NK cells. Soluble IL-15:IL15-Ra (IL-15/IL-15Ra) complexes have been shown to modulate immune responses via the IL-15RPyc complex, and the biological activity of IL- 15 has been shown to be increased 50-fold by administering it in a preformed complex of IL-15 and soluble IL-15Ra, which has an increased halflife compared to IL- 15 alone. This significant increase in the therapeutic efficacy of IL- 15 by pre-association with IL-15Ra has been demonstrated in murine tumor models (see, e.g., Han et al. (2011) Cytokine 56(3):804-810). IL-18

IL-18 induces the secretion of IFN-y by NK and CD8 ⁺ T-cells, enhancing their toxicity. IL-18 also activates macrophages and stimulates the development of Thl helper CD4 ⁺ T-cells. IL-18 has shown promising anti-tumor activity in several preclinical mouse models. For example, administration of recombinant IL- 18 (rIL-18) resulted in the regression of melanoma or sarcoma in syngeneic mice through the activation of CD4 ⁺ T-cells and/or NK cell-mediated responses. Other studies showed that IL- 18 anti -turn or effects were mediated by IFN-y, and involved anti angiogenic mechanisms. The combination of IL-18 with other cytokines, such as IL-12, or with co-stimulatory molecules, such as CD80, enhances the IL-18-mediated anti-tumor effects. Phase I clinical trials in patients with advanced solid tumors and lymphomas showed that IL- 18 administration was safe, and that it resulted in immune modulatory activity and in the increase of serum IFN-y and GM-CSF levels in patients, and in modest clinical responses. Clinical trials showed that IL- 18 can be combined with other anti-cancer therapeutic agents, such as monoclonal antibodies, cytotoxic drugs, or vaccines (see, e.g., Fabbi etal. (2015) J. Leukoc. Biol. 97:665-675; and Lee, S. and Margolin, K. (2011) Cancers 3:3856-3893).

It was found that an attenuated strain of Salmonella typhimurium, engineered to express IL-18, inhibited the growth of subcutaneous (S.C.) tumors or pulmonary metastases in syngeneic mice without any toxic effects following systemic administration. Treatment with this engineered bacterium induced the accumulation of T-cells, NK cells, and granulocytes in tumors, and resulted in the intratumoral production of cytokines (see, e.g., Fabbi et al. (2015) J. Leukoc. Biol. 97:665-675).

Chemokines

Chemokines are a family of small cytokines that mediate leukocyte migration to areas of injury or inflammation, and that are involved in mediating immune and inflammatory responses. Chemokines are classified into four subfamilies, based on the position of cysteine residues in their sequences, namely XC-, CC-, CXC-, and CX3C-chemokine ligands, or XCL, CCL, CXCL, and CX3CL. The chemokine ligands bind to their cognate receptors and regulate the circulation, homing, and retention of immune cells, with each chemokine ligand-receptor pair selectively regulating a certain type of immune cell. Different chemokines attract different leukocyte populations, and form a concentration gradient in vivo, with attracted immune cells moving through the gradient towards the higher concentration of chemokine (see, e.g., Argyle D. and Kitamura, T. (2018) Front. Immunol. 9:2629; and Dubinett etal. (2010) Cancer J. 16(4):325-335). Chemokines can improve the antitumor immune response by increasing the infiltration of immune cells into the tumor, and facilitating the movement of antigen-presenting cells (APCs) to tumor-draining lymph nodes, which primes naive T-cells and B cells (see, e.g., Lechner et al. (2011) Immunotherapy 3(11): 1317-1340). The immunostimulatory bacteria provided herein can be engineered to encode chemokines, including, but not limited to, CCL3, CCL4, CCL5, CXCL9, CXCL10, and CXCL11.

CCL3, CCL4, CCL5

CCL3, CCL4, and CCL5 share a high degree of homology, and bind to CCR5 (CCL3, CCL4, and CCL5) and CCR1 (CCL3 and CCL5) on several cell types, including immature DCs and T-cells, in both humans and mice. Therapeutic T-cells have been shown to induce chemotaxis of innate immune cells to tumor sites, via the tumor-specific secretion of CCL3, CCL4, and CCL5 (see, e.g., Dubinett et al. (2010) Cancer J. 16(4):325-335).

The induction of the T helper cell type 1 (Thl) response releases CCL3. In vivo and in vitro studies of mice have indicated that CCL3 is chemotactic for both neutrophils and monocytes; specifically, CCL3 can mediate myeloid precursor cell (MPC) mobilization from the bone marrow, and has MPC regulatory and stimulatory effects. Human ovarian carcinoma cells transfected with CCL3 showed enhanced T- cell infiltration and macrophages within the tumor, leading to an improved anti-tumor response, and indicated that CCL3-mediated chemotaxis of neutrophils suppressed tumor growth. DCs transfected with the tumor antigen human melanoma-associated gene (MAGE)-l that were recruited by CCL3 exhibited superior anti -tumor effects, including increased lymphocyte proliferation, cytolytic capacity, and survival, and decreased tumor growth, in a mouse model of melanoma. A combinatorial use of CCL3 with an antigen-specific platform for MAGE-1 has also been used in the treatment of gastric cancer. CCL3 production by CT26, a highly immunogenic murine colon tumor, slowed in vivo tumor growth; this process was driven by the CCL3- dependent accumulation of natural killer (NK) cells, and thus, IFNy, resulting in the production of CXCL9 and CXLC10 (see, e.g., Allen et al. (2017) Oncoimmunology 7(3):el393598; and Schaller et al. (2017) Expert Rev. Clin. Immunol. 13(11):1049- 1060). CCL3 has been used as an adjuvant for the treatment of cancer. Administration of a CCL3 active variant, ECI301, after radiofrequency ablation in mouse hepatocellular carcinoma increased tumor-specific responses, and this mechanism was further shown to be dependent on the expression of CCR1. CCL3 has also shown success as an adjuvant in systemic cancers, whereby mice vaccinated with CCL3 and IL-2 or granulocyte-macrophage colony-stimulating factor (GM-CSF), in a model of leukemia/lymphoma, exhibited increased survival (see, e.g., Schaller et al. (2017) Expert Rev. Clin. Immunol. 13(11): 1049-1060).

CCL3 and CCL4 play a role in directing CD8 ⁺ T-cell infiltration into primary tumor sites in melanoma and colon cancers. Tumor production of CCL4 leads to the accumulation of CD103 ⁺ DCs; suppression of CCL4 through a WNT/p-catenin- dependent pathway prevented CD103 ⁺ DC infiltration of melanoma tumors (see, e.g., Spranger et al. 2015) Nature 523(7559):231-235). CCL3 was also shown to enhance CD4 ⁺ and CD8 ⁺ T-cell infiltration to the primary tumor site in a mouse model of colon cancer (see, e.g., Allen et al. (2017) Oncoimmunology 7(3):el393598).

The binding of CCL3 or CCL5 to their receptors (CCR1 and CCR5), moves immature DCs, monocytes, and memory and T effector cells from the circulation into sites of inflammation or infection. For example, CCL5 expression in colorectal tumors contributes to T lymphocyte chemoattraction and survival. CCL3 and CCL5 have been used alone or in combination therapy to induce tumor regression and immunity in several preclinical models. For example, studies have shown that the subcutaneous injection of Chinese hamster ovary cells genetically modified to express CCL3, resulted in tumor inhibition and neutrophilic infiltration. In another study, a recombinant oncolytic adenovirus expressing CCL5 (Ad-RANTES-EIA) resulted in primary tumor regression, and blocked metastasis in a mammary carcinoma murine model (see, e.g., Lechner et al. (2011) Immunotherapy 3(11): 1317-1340).

In a translational study of colorectal cancer, CCL5 induced an “antiviral response pattern” in macrophages. As a result of CXCR3 -mediated migration of lymphocytes at the invasive margin of liver metastases in colorectal cancer, CCL5 is produced. Blockade of CCR5, the CCL5 receptor, results in tumor death, driven by macrophages producing IFN and reactive oxygen species. While macrophages are present in the tumor microenvironment, CCR5 inhibition induces a phenotypic shift from an M2 to an Ml phenotype. CCR5 blockade also leads to clinical responses in colorectal cancer patients (see, e.g., Halama etal. (2016) Cancer Cell 29(4):587-601). CCL3, CCL4, and CCL5 can be used for treating conditions, including lymphatic tumors, bladder cancer, colorectal cancer, lung cancer, melanoma, pancreatic cancer, ovarian cancer, cervical cancer, or liver cancer (see, e.g., U.S. Patent Publication No. US 2015/0232880; and International Application Publication Nos. WO 2015/059303, WO 2017/043815, WO 2017/156349, and WO 2018/191654).

CXCL9, CXCL10, CXCL11

CXCL9 (MIG), CXCL10 (IP 10), and CXCL11 (IT AC) are induced by the production of IFN-y. These chemokines bind CXCR3, preferentially expressed on activated T-cells, and function both angiostatically, and in the recruitment and activation of leukocytes. Prognosis in colorectal cancer is strongly correlated to tumor-infiltrating T-cells, particularly Thl and CD8 ⁺ effector T-cells; high intratumoral expression of CXCL9, CXCL10 and CXCL11 is indicative of good prognosis. For example, in a sample of 163 patients with colon cancer, those with high levels of CXCL9 or CXCL11 showed increased post-operative survival, and patients with high CXC expression had significantly higher numbers of CD3 ⁺ T-cells, CD4 ⁺ T-helper cells, and CD8 ⁺ cytotoxic T-cells. In liver metastases of colorectal cancer patients, CXCL9 and CXCL10 levels were increased at the invasive margin, and correlated with effector T-cell density. The stimulation of lymphocyte migration via the action of CXCL9 and CXCL10 on CXCR3 leads to the production of CCL5 at the invasive margin (see, e.g., Halama et al. (2016) Cancer Cell 29(4):587-601; and Kistner et al. (2017) Oncotarget 8(52):89998-90012).

In vivo, CXCL9 functions as a chemoattractant for tumor-infiltrating lymphocytes (TILs), activated peripheral blood lymphocytes, natural killer (NK) cells, and Thl lymphocytes. CXCL9 also is critical for T-cell-mediated suppression of cutaneous tumors. For example, when combined with systemic IL-2, CXCL9 has been shown to inhibit tumor growth via the increased intratumoral infiltration of CXCR3 ⁺ mononuclear cells. In a murine model of colon carcinoma, a combination of the huKSl/4-IL-2 fusion protein with CXCL9 gene therapy achieved a superior antitumor effect and prolonged lifespan through the chemoattraction and activation of CD8 ⁺ and CD4 ⁺ T lymphocytes (see, e.g., Dubinett et al. (2010) Cancer J. 16(4):325- 335; and Ruehlmann et al. (2001) Cancer Res. 61(23):8498-8503).

CXCL10, produced by activated monocytes, fibroblasts, endothelial cells and keratinocytes, is chemotactic for activated T-cells, and can act as an inhibitor of angiogenesis in vivo. Expression of CXCL10 in colorectal tumors has been shown to contribute to cytotoxic T lymphocyte chemoattraction and longer survival. The administration of immunostimulatory cytokines, such as IL-12, has been shown to enhance the anti -tumor effects generated by CXCL10. A dendritic cell (DC) vaccine primed with a tumor cell lysate, and transfected with CXCL10 had increased immunological protection and effectiveness in mice; the animals showed a resistance to a tumor challenge, a slowing of tumor growth, and longer survival time. In vivo and in vitro studies in mice using the CXCLIO-mucin-GPI fusion protein resulted in tumors with higher levels of recruited NK cells compared to tumors not treated with the fusion protein. Interferons (which can be produced by plasmacytoid dendritic cells; these cells are associated with primary melanoma lesions and can be recruited to a tumor site by CCL20) can act on tumor DC subsets, for example, CD103 ⁺ DCs, which have been shown to produce CXCL9/10 in a mouse melanoma model, and have been associated with CXCL9/10 in human disease. CXCL10 also has shown higher expression in human metastatic melanoma samples relative to primary melanoma samples. Therapeutically, adjuvant IFN-a melanoma therapy upregulates CXCL10 production, whereas the chemotherapy agent cisplatin induces CXCL9 and CXCL10 (see, e.g., Dubinett etal. (2010) Cancer J. 16(4):325-335; Kuo etal. (2018) Front. Med. (Lausanne) 5:271; Li etal. (2007) Scand. J. Immunol. 65(1):8-13; and Muenchmeier et al. (2013) PLoS One 8(8):e72749).

CXCL10/11 and CXCR3 expression has been established in human keratinocytes derived from basal cell carcinomas (BCCs). CXCL11 also is capable of promoting immunosuppressive indoleamine 2,3 -dioxygenase (IDO) expression in human basal cell carcinoma, as well as enhancing keratinocyte proliferation, which could reduce the anti-tumor activity of any infiltrating CXCR3 ⁺ effector T-cells (see, e.g., Kuo et al. (2018) Front. Med. (Lausanne) 5:271).

CXCL9, CXCL10 and CXCL11 can be encoded in oncolytic viruses for treating cancer (see, e.g., U.S. Patent Publication No. 2015/0232880; and International Application Publication No. WO 2015/059303). Pseudotyped oncolytic viruses or a genetically engineered bacterium encoding the gene for CXCL10 also can be used to treat cancer (see, e.g., International Application Publication Nos. WO 2018/006005 and WO 2018/129404). b. Co-Stimulatory Molecules Co-stimulatory molecules enhance the immune response against tumor cells, and co-stimulatory pathways are inhibited by tumor cells to promote tumorigenesis. The immunostimulatory bacteria herein can be engineered to express co-stimulatory molecules, such as, for example, CD40, CD40L, 4-1BB, 4-1BBL, 4-1BBL with a deletion of the cytoplasmic domain (4-lBBLAcyt), 4-1BBL with a truncated cytoplasmic domain, 0X40 (CD134), OX40L (CD252), other members of the TNFR superfamily e.g., CD27, CD27 ligand, GITR, CD30, Fas receptor, TRAIL-R, TNF-R, HVEM, and RANK), B7, CD80, CD86, ICOS, ICOS ligand (B7RP1), and CD28. Additionally, the immunostimulatory bacteria can encode and express truncated co- stimulatory molecules (e.g., 4-1BBL, CD80, CD86, CD27L, B7RP1, OX40L), with a full or partial (complete, or truncated, or modified to ensure proper orientation when expressed in a cell) cytoplasmic domain deletion, for expression on an antigen- presenting cell (APC). The gene product with a truncated cytoplasmic domain, including a full deletion, exhibits constitutive immunostimulatory signaling to a T-cell through co-stimulatory receptor engagement, and is unable to counter-regulatory signal to the APC due to a truncated or deleted (or otherwise modified as described herein) cytoplasmic domain. The truncation is sufficient to provide the signaling, and for the modified co-stimulatory molecule to be unable to counter-regulatory signal to the APC. The complete or partial deletion of the cytoplasmic domain of a co- stimulatory molecule, as described herein, potentiates the activation of the co- stimulatory molecule, without the immunosuppressive reverse signaling. The partial deletion (or truncation) of the cytoplasmic domain is a sufficient deletion to achieve these effects, without affecting the expression of the co-stimulatory molecule, or the orientation of the expressed co-stimulatory molecule.

The co-stimulatory molecules also can be modified to eliminate or reduce the immunosuppressive intracellular/reverse signaling by modifications to the amino acids in the cytoplasmic domain, including insertions, deletions, and/or replacements. In particular, the co-stimulatory molecules are modified by modification, such as by replacement, of cytoplasmic domain phosphorylation sites. For example, replacing one or more Ser residues at an appropriate locus or loci, such as, for human 4-1BBL, Ser5 and Ser8, with a residue that reduces or eliminates reverse signaling.

The immunostimulatory bacteria herein also can be engineered to express agonistic antibodies against co-stimulatory molecules (e.g., 4-1BBL) to enhance the anti-tumor immune response. TNF Receptor Superfamily

The TNF superfamily of ligands (TNFSF) and their receptors (TNFRSF) are involved in the proliferation, differentiation, activation and survival of tumor and immune effector cells. Members of this family include CD30, Fas-L, TRAIL-R, and TNF-R, which induce apoptosis, and CD27, OX40L, CD40L, GITR-L, and 4-1BBL, which regulate B and T-cell immune responses. Other members include herpesvirus entry mediator (HVEM). The expression of TNFSF and TNFRSF by the immunostimulatory bacteria herein can enhance the anti-tumor immune response. It has been shown, for example, that the expression of 4-1BBL in murine tumors enhances immunogenicity, and that intratumoral injection of dendritic cells (DCs) with increased expression of OX40L can result in tumor rejection in murine models. Studies have also shown that injection of an adenovirus expressing recombinant GITR into B 16 melanoma cells promotes T-cell infiltration and reduces tumor volume. Stimulatory antibodies against molecules such as 4-1BB, 0X40 and GITR also can be encoded by the immunostimulatory bacteria to stimulate the immune system. For example, agonistic anti -4- IBB monoclonal antibodies have been shown to enhance anti-tumor CTL responses, and agonistic anti-OX40 antibodies have been shown to increase anti-tumor activity in transplantable tumor models. Additionally, agonistic anti-GITR antibodies have been shown to enhance anti-tumor responses and immunity (see, e.g., Lechner etal. (2011) Immunotherapy 3(11): 1317-1340; and Peggs etal. (2009) Clinical and Experimental Immunology 157:9-19).

CD40 and CD40L

CD40, which is a member of the TNF receptor superfamily, is expressed by APCs and B cells, while its ligand, CD40L (CD 154), is expressed by activated T- cells. Interaction between CD40 and CD40L stimulates B cells to produce cytokines, resulting in T-cell activation and tumor cell death. Studies have shown that anti-tumor immune responses are impaired with reduced expression of CD40L on T-cells or CD40 on dendritic cells. CD40 is expressed on the surface of several B-cell tumors, such as follicular lymphoma, Burkitt lymphoma, lymphoblastic leukemia, and chronic lymphocytic leukemia, and its interaction with CD40L has been shown to increase the expression of B7-1/CD80, B7-2/CD86, and human leukocyte antigen (HLA) class II molecules in the CD40 ⁺ tumor cells, as well as enhance their antigen-presenting abilities. Transgenic expression of CD40L in a murine model of multiple myeloma resulted in the induction of CD4 ⁺ and CD8 ⁺ T-cells, local and systemic anti-tumor immune responses, and reduced tumor growth. Anti-CD40 agonistic antibodies also induced anti-tumor T-cell responses (see, e.g., Marin-Acevedo etal. (2018) Journal of Hematology & Oncology 11:39; Dotti etal. (2002) Blood 100(l):200-207; and Murugaiyan et al. (2007) J. Immunol. 178:2047-2055).

4-1BB and 4-1BBL

4-1BB (CD137) is an inducible co-stimulatory receptor that is expressed primarily by T-cells and NK cells; it binds its ligand 4-1BBL, that is expressed on APCs, including DCs, B-cells, and monocytes, to trigger immune cell proliferation and activation. 4- IBB results in longer and more widespread responses of activated T- cells. Anti-4-lBB agonists and 4-1BBL fusion proteins have been shown to increase immune-mediated anti-tumor activity, for example, against sarcoma and mastocytoma tumors, mediated by CD4 ⁺ Thl and tumor-specific CTL activity (see, e.g., Lechner et al. (2011) Immunotherapy 3(11): 1317-1340; and Marin-Acevedo et al. (2018) Journal of Hematology & Oncology 11:39). 4-1BBL is negatively regulated by its cytoplasmic signaling domain. In the late-phase of 4-1BBL ligation on macrophages to T-cells, reverse signaling of the 4-1BBL cytoplasmic domain induces surface translocation of 4-1BBL to bind to form a signaling complex with TLR4. This induces high levels of TNF-a, comparable to LPS activation of TLR4, that leads to immunosuppression of the adaptive immune response (see, e.g., Ma et al. (2013) Set. Signaling 295(6) : 1 - 11). Deletion of the cytoplasmic domain of 4-1BBL, as described herein, potentiates the activation of 4-1BBL without the immunosuppressive reverse signaling.

0X40 and OX40L

0X40 (CD134) is a member of the TNF receptor superfamily that is expressed on activated effector T-cells, while its ligand, OX40L is expressed on APCs, including DCs, B cells and macrophages, following activation by TLR agonists and CD40-CD40L signaling. OX40-OX40L signaling results in the activation, potentiation, proliferation and survival of T-cells, as well as the modulation of NK cell function and inhibition of the suppressive activity of Tregs. Signaling through 0X40 also results in the secretion of cytokines (IL-2, IL-4, IL-5, and IFN-y), boosting Thl and Th2 cell responses. The recognition of tumor antigens by tumor-infiltrating lymphocytes (TILs) results in increased expression of 0X40 by the TILs, which has been correlated with improved prognosis. Studies have demonstrated that treatment with anti-OX40 agonist antibodies or Fc-OX40L fusion proteins results in enhanced tumor-specific CD4 ⁺ T-cell responses and increased survival in murine models of melanoma, sarcoma, colon carcinoma, and breast cancer, while Fc-OX40L incorporated into tumor cell vaccines protected mice from subsequent challenge with breast carcinoma cells (see, e.g., Lechner et al. (2011) Immunotherapy 3(11): 1317- 1340; and Marin-Acevedo et al. (2018) Journal of Hematology & Oncology 11 :39).

B7-CD28 Family

CD28 is a co-stimulatory molecule expressed on the surface of T-cells that acts as a receptor for B7-1 (CD80) and B7-2 (CD86), which are co-stimulatory molecules expressed on antigen-presenting cells. CD28-B7 signaling is required for T-cell activation and survival, and for the prevention of T-cell anergy, and results in the production of interleukins, such as IL-6.

Optimal T-cell priming requires two signals: (1) T-cell receptor (TCR) recognition of MHC -presented antigens, and (2) co-stimulatory signals resulting from the ligation of T-cell CD28 with B7-1 (CD80) or B7-2 (CD86) expressed on APCs. Following T-cell activation, CTLA-4 receptors are induced, which then outcompete CD28 for binding to B7-1 and B7-2 ligands. Antigen presentation by tumor cells is poor due to their lack of expression of co-stimulatory molecules, such as B7-1/CD80 and B7-2/CD86, resulting in a failure to activate the T-cell receptor complex. As a result, upregulation of these molecules on the surfaces of tumor cells can enhance their immunogenicity. Immunotherapy of solid tumors and hematologic malignancies has been successfully induced by B7, for example, via tumor cell expression of B7, or soluble B7-immunoglobulin fusion proteins. The viral-mediated tumor expression of B7, in combination with other co-stimulatory ligands, such as ICAM-3 and LFA-3, has been successful in preclinical and clinical trials for the treatment of chronic lymphocytic leukemia and metastatic melanoma. Additionally, soluble B7 fusion proteins have demonstrated promising results in the immunotherapy of solid tumors as single agent immunotherapies (see, e.g., Lechner etal. (2011) Immunotherapy 3(11): 1317-1340; and Dotti etal. (2002) Blood 100(l):200-207).

2. Constitutively Active Proteins that Stimulate the Immune Response and/or Type I IFN, Non-Human STING Proteins, STING Chimeras, and Modified Forms

Type I interferons (IFNs; also referred to as interferon type 1), include IFN-a and IFN-P, and are pleiotropic cytokines with antiviral, anti-tumor, and immunoregulatory activities. IFN-p is produced by most cell types; IFN-a primarily is produced by hematopoietic cells, particularly plasmacytoid dendritic cells. Type I IFNs are produced following the sensing of pathogen-associated molecular patterns (PAMPs) by pattern recognition receptors (PRRs). They are involved in the innate immune response against pathogens, mainly viral, and are potent immunomodulators that promote antigen presentation, mediate dendritic cell (DC) maturation, activate cytotoxic T lymphocytes (CTLs), natural killer (NK) cells and macrophages, and activate the adaptive immune system by promoting the development of high-affinity antigen-specific T-cell and B-cell responses and immunological memory.

Type I IFNs exhibit anti-proliferative and pro-apoptotic effects on tumors and have anti-angiogenic effects on tumor neovasculature. They induce the expression of MHC class I molecules on tumor cell surfaces, increase the immunogenicity of tumor cells, and activate cytotoxicity against them. Type I IFN has been used as a therapeutic for the treatment of cancers and viral infections. For example, IFN-a (sold under the trademark Intron®/Roferon®-A) is approved for the treatment of hairy cell leukemia, malignant melanoma, AIDS-related Kaposi’s sarcoma, and follicular nonHodgkin’s lymphoma; it also is used in the treatment of chronic myelogenous leukemia (CML), renal cell carcinoma, neuroendocrine tumors, multiple myeloma, non-follicular non-Hodgkin’s lymphoma, desmoid tumors, and cutaneous T-cell lymphoma, although use is limited due to systemic immunotoxicity (see, e.g., Ivashkiv and Donlin (2014) Nat. Rev. Immunol. 14(l):36-49; Kalliolias and Ivashkiv (2010) Arthritis Research & Therapy 12(Suppl 1):S1; and Lee, S. and Margolin, K. (2011) Cancers 3:3856-3893).

Expression of type I interferons in tumors and the tumor microenvironment is among the immune responses that the immunostimulatory bacteria herein are designed to evoke. Inducing or evoking type I interferon provides anti-tumor immunity for the treatment of cancer. a. Constitutive STING Expression and Gain-of-Function Mutations

The induction of type I IFNs, proinflammatory cytokines and chemokines is necessary for mounting an immune response that prevents or inhibits infection by viral pathogens. This response also can be effective as an anti -turn or agent. The immunostimulatory bacteria provided herein encode proteins that constitutively induce type I IFNs. Among these proteins are those that occur in individuals with various diseases or disorders that involve the over-production of immune response modulators. For example, over-production or excessive production, or defective negative regulation of type I IFNs and pro-inflammatory cytokines, can lead to undesirable effects, such as inflammatory and autoimmune diseases. Disorders involving the overproduction, generally chronic, of type I IFNs, are referred to as interferonopathies (see, e.g., Lu and MacDougall (2017) Front. Genet. 8: 118; and Konno et al. (2018) Cell Reports 23 : 1112-1123). Disorders and clinical phenotypes associated with type I interferonopathies include Aicardi-Goutieres syndrome (AGS), STING-associated vasculopathy with onset in infancy (SAVI), Singleton-Merten syndrome (SMS), atypical SMS, familial chilblain lupus (FCL), systemic lupus erythematosus (SLE), bilateral striatal necrosis (BSN), cerebrovascular disease (CVD), dyschromatosis symmetrica hereditaria (DSH), spastic paraparesis (SP), X- linked reticulate pigmentary disorder (XLPDR), proteasome-associated auto- inflammatory syndrome (PRAAS), intracranial calcification (ICC), Mendelian susceptibility to mycobacterial disease (MSMD), and spondyloenchondrodysplasia (SPENCD) (see, e.g, Rodero et a/. (2016) Exp. Med. 213(12):2527-2538). These phenotypes are associated with particular genotypes, involving mutations in genes that lead to constitutive activities of products involved in the induction of type I IFNs.

The sustained activation of interferon signaling can be due to: 1) loss-of- function mutations leading to increased cytosolic DNA (e.g., mutations in TREX1 and SAMHD1), or increased cytosolic RNA/DNA hybrids (e.g., mutations in RNASEH2A, RNASEH2B, RNASEH2C, and POLA1),' 2) loss-of-function mutations resulting in a defect in RNA editing and abnormal sensing of self-nucleic acid RNA species in the cytosol (e.g., mutations m A DARI)'. 3) gain-of-function mutations leading to constitutive activation of cytosolic IFN signaling pathways/increased sensitivity to cytosolic nucleic acid ligands e.g., mutations in RIG-I, MDA5 and STING); 4) loss- of-function mutations leading to aberrant RNA signaling via MAVS caused by a disturbance of the unfolded protein response e.g., mutations in SKIV2L),' 5) loss-of- function mutations in molecules responsible for limiting IFN receptor (IFNAR1/2) signaling, leading to uncontrolled IFN-stimulated gene (ISG) production e.g., mutations in USP18 and ISG 15), 6) proteasomal dysfunction, leading to increased IFN signaling through an unknown mechanism (e.g., mutations in PSMA3, PSMB4 and PSMB8),' and 7) loss-of-function mutations in TRAP A4CP5 and Clq, where the mechanisms leading to type I IFN signaling remain unclear (see, e.g., Rodero et al. (2016) J. Exp. Med. 2I3(I2):2527-2538). Of interest herein are mutations that lead to gain-of-function (GOF). There are known mutations in STING, MDA5 and RIG-I, that are associated with constitutive activation of the encoded proteins, and/or enhanced sensitivity or increased affinity or binding to endogenous ligands. GOF mutations in STING, for example, are linked to SAVI and FCL; GOF mutations in MDA5 are linked to AGS and SMS; and GOF mutations in RIG-I are linked to atypical SMS.

TMEM1 73 STING Alleles

Stimulator of interferon genes (STING) is encoded by the transmembrane protein 173 (TMEM173) gene, which is a ~7 kb-long gene. The human TMEM173 gene is characterized by significant heterogeneity and population stratification of alleles. The most common human TMEM173 allele is referred to as R232 (referencing the amino acid present at residue 232; see, e.g., SEQ ID N0s:305-309, setting forth the sequences of various human TMEM173 alleles). More than half the American population is noiR232/R232. The second most common allele is R7IH-G230A- R293Q (HAQ). Other common alleles include AQ (G230A-R293Q), Q (Q293) and R232H (named REF after the reference STING allele first identified and catalogued in the database by Glen Barber).

R232/R232 is the most common genotype in Europeans, while HAQ/R232 is the most common genotype in East Asians. Africans have no HAQ/HAQ genotypes, but have the Q allele, and ~4% of Africans are AQ/AQ, which is absent in other ethnic populations (see, e.g., Patel and lin (2018) Genes & Immunity, doi: 10.1038/s41435- 018-0029-9). The REF, AQ and Q alleles are highly refractory to bacterially-derived CDNs, such as 3’3’ c-di-GMP (see, e.g., Corrales etal. (2015) Cell Reports IT.1018- 1030).

STING Gain-of-Function Mutations

Several activating or gain-of-function (GOF) mutations in TMEM173, the gene for STING, inherited and de novo, have been linked to the rare auto- inflammatory disease SAVI (STING-associated vasculopathy with onset in infancy). SAVI is an autosomal dominant disease and is characterized by systemic inflammation, interstitial lung disease, cutaneous vasculitis, and recurrent bacterial infection. SAVI with de novo TMEM173 mutations typically is characterized by an early-onset (< 8 weeks) and severe phenotype, while familial mutations result in late- onset (teens to adults) and milder clinical symptoms. Inherited TMEM173 activating mutations include G166E and V155M, whereas de novo mutations include N154S, V155M, V147M, V147L, C206Y, R284G, R281Q, and S102P/F279L (see, e.g, Patel and Jin (2019) Genes & Immunity 20:82-89). Other activating TMEM173 mutations that have been identified include R284M, R284K, R284T, E316Q, and R375A (see, e.g., U.S. Application Publication No. 2018/0311343). Another gain-of-function mutation in TMEM173 is R284S, which results in a highly constitutively active STING, and was found to trigger innate immune signaling in the absence of activating CDNs, leading to chronic production of pro-inflammatory cytokines (see, e.g., Konno et al. (2018) Cell Reports 23 : 1112- 1123).

TMEM173 mutations, such as N154S, V155M and V147L, and/or any of the mutations listed in the table below, singly or in any combination, with these and any other such mutations, such as N154S/R284G, result in a gain-of-function STING that is constitutively active and does not require, or is hypersensitive to, ligand stimulation, leading to chronic activation of the STING-interferon pathway. This has been demonstrated (see, e.g., Liu et al. (2014) A. Engl. J. Med. 371:507-518). Constructs of mutated TMEM173 (with each of the replacements V147L, N154S, V155M, and the loss-of-function mutant V155R), and non-mutated TMEM173, were transfected into STING-negative HEK293T cells, and stimulated with the STING ligand, cGAMP. Cells transfected with the N154S, V155M and V147L mutants exhibited highly elevated IFNB1 (the gene encoding IFN- ) reporter activity, which was not significantly boosted by stimulation with the STING ligand cGAMP. Cells that were transfected with the loss-of-function mutant (V155R), non-mutated TMEM173, or control plasmid, had no significant baseline activation. Stimulation with cGAMP resulted in a response in a dose-dependent manner in cells with nonmutated TMEM173, and resulted in a minimal response only at the highest cGAMP concentration, in cells expressing the loss-of-function mutant (see, e.g., Liu et al. (2014) N. Engl. J. Med. 371 :507-518). These results show that the activating TMEM173 mutations result in constitutive activation of STING, even in the absence of stimulation by cGAMP.

G207E is another gain-of-function STING mutation that causes alopecia, photosensitivity, thyroid dysfunction, and SAVLfeatures. The G207E mutation causes constitutive activation of inflammation-related pathways in HEK cells, as well as aberrant interferon signature and inflammasome activation in patient peripheral blood mononuclear cells (PBMCs). Using STING variants with the R232 or H232 allele and the GOF mutation G207E, it was shown that after stimulation with CDN, the R232 + G207E vanant resulted in slight increases of activity in the IFN-p and STAT1/2 pathways, while with the H232 + G207E variant, IFN-P levels remained constant, and STAT1/2 showed diminished activity. Both variants showed similar STAT3 and NF- KB pathway activation following stimulation. These results show that the residue R at position 232 is important for cGAMP binding and IFN induction, and show that G207E mutants result in constitutive activation of STING signaling pathways and ligand-dependent hyperactivation of the NF-KB pathway. Patients with the R232 allele and the G207E mutation had more severe disease; this polymorphism strengthens the constitutive activation of the mutant STING, leading to the overexpression of downstream targets, such as IFN, IL1- , and IL- 18 (see, e.g., Keskitalo et al. (2018), available from: doi.org/10.1101/394353).

67 amino acids in murine STING (SEQ ID NO:369) were mutated (see, Burdette et al. (2011) Nature 478(7370):515-518), either individually or in groups, to identify amino acids involved in cyclic di-GMP (c-di-GMP) binding and/or IFN induction. Among the mutants identified were hyperactive mutants R196A/D204A, S271A/Q272A, R309A/E315A, E315A, E315N, E315Q, and S271A (corresponding to R197A/D205A, S272A/Q273A, R310A/E316A, E316A, E316N, E316Q, and S272A, respectively, with reference to the sequence of human STING as set forth in SEQ ID N0s:305-309), that spontaneously induced IFN at low levels of transfection and did not respond to c-di-GMP, and the mutants R374A, R292A/T293A/E295A/E299A, D230A, R231A, K235A, Q272A, S357A/E359A/S365A, D230A/R231A/K235A/R237A, and R237A (corresponding to R375A, R293A/T294A/E296A (there is no equivalent to E299 in human STING), D231A, R232A, K236A, Q273A, S358A/E360A/S366A, D231A/R232A/K236A/R238A, and R238A, respectively, with reference to human STING, as set forth in SEQ ID N0s:305-309), that induced IFN when overexpressed but did not respond to c-di-GMP. These alleles can still respond to the endogenous CDN 2’3’ c-di-GAMP, as it was later discovered that some human STING mutations have low affinity for the 3’3’ CDNs produced by bacteria, such as c-di-GMP (see, e.g., Corrales et al. (2015) Cell Reports 11:1018-1030).

The immunostimulatory bacteria provided herein that encode these proteins with gain-of-function mutations exploit the constitutive activation of these proteins to increase production of type I IFNs and pro-inflammatory cytokines. Tumor-targeting immunostimulatory bacteria are provided herein that encode STING, IRF3, IRF5, IRF7, MDA5, and/or RIG-I, with gain-of -function mutations. The immunostimulatory bacteria increase the production of type I IFN-mediated cytokines and chemokines in the tumor microenvironment, potentiating the anti-tumor immune response and improving the therapeutic efficacy of the immunostimulatory bacteria. The gene encoding STING is referred to as TMEM173, the gene encoding MDA5 is IFIH1, and the gene encoding RIG-I is DDX58. There are numerous alleles for each gene, and known mutations that can occur in genes with any of the alleles, resulting in gain-of- function or constitutive activation. The mutations listed below can occur singly, or can be used in any combination. Other mutations that result in gain-of-function can be identified by routine screening/mutation protocols. The table below lists exemplary gain-of-function mutations in each of STING/TMEM173 (SEQ ID NOsG05-309), MDA5/IFIH1 (SEQ ID NOG 10), RIG-I/DDX58 (SEQ ID NOG 11), IRF3 (SEQ ID NOG 12), and IRF7 (SEQ ID NOG 13). Other mutations, such as deletion of, or replacement of, a phosphorylation site or sites, such as S324/L325/S326 — ► S324A/L325/S326A in STING, and other replacements to eliminate a phosphorylation site to reduce nuclear factor-KB (NF-KB) signaling in STING, or other proteins that employ such signaling, also can be introduced.

The resulting proteins can be encoded in the immunostimulatory bacteria provided herein. The proteins are encoded on plasmids in the immunostimulatory bacteria.

Administering nucleic acids encoding wild-type STING can induce an immune response; the administration of gain-of-function STING mutants, with constitutive activity as provided herein, in tumor-targeted immunostimulatory bacteria, leads to a more potent immune response and more effective anti-cancer therapeutic. The enhanced immune response by the tumor-targeted administration of constitutively active STING, or other such modified DNA/RNA sensors, such as gain- of-function mutants of MDA5, RIG-I, IRF3, or IRF7, as provided herein, provides a therapeutically more effective anti-cancer treatment. For example, as described herein, modifying the immunostimulatory bacteria so that they do not infect epithelial cells, but retain the ability to infect phagocytic cells, including tumor-resident immune cells, effectively targets the immunostimulatory bacteria to the tumor microenvironment, improving therapeutic efficiency and preventing undesirable systemic immune responses. These tumor-targeted bacteria are engineered to encode gain-of-function STING, MDA5, RIG-I, IRF3, or IRF7 mutants, which are constitutively active, for example, even in the absence of ligand stimulation, providing a potent type I IFN response to improve the anti-cancer immune response in the tumor microenvironment.

Thus, for example, the administration of constitutively activated STING can provide an alternative means to boost STING signaling for the immunotherapeutic treatment of cancer. In certain embodiments, the tumor-targeting immunostimulatory bacteria provided herein can be modified to encode STING/TMEM173 (SEQ ID NOs: 305-309) with gain-of-function mutations, selected from S102P, V147L, V147M, N154S, V155M, G166E, R197A, D205A, R197A/D205A, C206Y, G207E, D231A, R232A, K236A, R238A, D231A/R232A/K236A/R238A, S272A, Q273A,

S272A/Q273A, F279L, S102P/F279L, R281Q, R284G, R284S, R284M, R284K, R284T, R293A, T294A, E296A, R293A/T294A/E296A, R310A, E316A, E316N, E316Q, R310A/E316A, S324A/S326A, S358A, E360A, S366A, S358A/E360A/S366A, and R375A, as well as conservative mutations thereof. In addition, combinations of STING gain-of-function mutations, such as N154S/R284G, significantly increase STING signaling, compared to the single mutation counterparts. These mutations result in constitutive activity of STING.

Table of Exemplary Gain-Of-Function Mutants

Gain-of-Function Mutations Resulting in the Persistent Expression of Type I IFN

Amino acid residues R197, D205, R310, R293, T294, E296, S272, Q273, E316,

D231, R232, K236, S358, E360, S366, and R238, with reference to the sequence of human STING, as set forth in any of SEQ ID N0s:305-309, correspond to amino acid residues R196, D204, R309, R292, T293, E295, S271, Q272, E315, D230, R231, K235, S357, E359, S365, and R237, respectively, of the murine STING, as set forth in SEQ ID NO:369. Also included are conservative substitutions of each of the replacements (see, e.g., the Table in the Definitions section, listing exemplary conservative mutations for each amino acid). b. Constitutive IRF3 Expression and Gain-of-Function Mutations

IRF3 (interferon regulatory factor 3, or IRF-3) and IRF7 (or IRF-7) are key activators of type I IFN genes. Following virus-induced C-terminal phosphorylation (by TBK1), activated IRF3 and IRF7 form homodimers, translocate from the cytoplasm to the nucleus, and bind to IFN-stimulated response elements (ISREs) to induce type I IFN responses. IRF3 is expressed constitutively in unstimulated cells, and exists as an inactive cytoplasmic form, while IRF7 is not constitutively expressed in cells, and is induced by IFN, lipopolysaccharide and virus infection. Overexpression of IRF3 significantly increases the virus-mediated expression of type I IFN genes, resulting in the induction of an antiviral state. IRF3 activation also has been shown to up-regulate the transcription of the CC-chemokine RANTES (CCL5) following viral infection (see, e.g., Lin etal. (1999) Mol. Cell Biol. 19(4):2465-2474).

Residues S385, S386, S396, S398, S402, T404, and S405 in the C-terminal domain of IRF3 are phosphorylated after virus infection, inducing a conformational change that results in the activation of IRF3. IRF3 activation is induced, not only by viral infection, but also by lipopolysaccharide (LPS) and poly(LC). Of the seven residues that can be phosphorylated in the C-terminal cluster of IRF3, a single point mutation, S396D, is sufficient for the generation of a constitutively active form of IRF3. IRF3(S396D) enhances the transactivation of IFNal, IFN-P, and RANTES promoters by 13-, 14-, and 11-fold, respectively, compared to wild-type IRF3. Another mutant, IRF3(S396D/S398D), enhances the transactivation of IFNal, IFN-P, and RANTES promoters by 13-, 12-, and 12-fold, respectively, compared to wild-type IRF3. Another constitutively active mutant of IRF3 is IRF3(5D), in which the serine or threonine residues at positions 396, 398, 402, 404, and 405 are replaced by phosphomimetic aspartic acid residues (IRF3(S396D/S398D/S402D/T404D/S405D)). Similar gain-of-function mutations, leading to constitutive activity of immune response mediators, such as induction of type I interferon, can be achieved by mutating serine residues to phosphomimetic aspartic acid in other proteins, such as RIG-I, MDA5 and STING, that are in immune response signaling pathways.

IRF3(5D) displays constitutive DNA binding and transactivation activities, dimer formation, association with the transcription coactivators p300 (also called EP300, or El A binding protein p300)/ CBP (also known as CREB-binding protein, or CREBBP), and nuclear localization. Its transactivation activity is not induced further by virus infection. IRF3(5D) is a very strong activator of IFN-P and ISG-15 gene expression; IRF3(5D) alone stimulates IFN-P expression as strongly as virus infection, and enhances transactivation of IFNal, IFN-P, and RANTES promoters by 9-fold, 5.5-fold, and 8-fold, respectively, compared to wild-type IRF3 (see, e.g., Lin et al. (2000) 7 Biol. Chem. 275(44):34320-34327; Lin et al. (199S)Mol. Cell Biol. 18(5):2986-2996; and Servant et al. (2003) 7 Biol. Chem. 278(11):9441-9447). Any of positions S385, S386, S396, S398, S402, T404, and S405 can be mutated, alone or in combination, to produce constitutively active IRF3 mutants in the immunostimulatory bacteria provided herein. c. Non-Human STING Proteins, and Variants Thereof with Increased or Constitutive Activity, and STING Chimeras, and Variants Thereof with Increased or Constitutive Activity

As discussed above, cytosolic double-stranded DNA (dsDNA) stimulates the production of type I interferon (IFN) through the endoplasmic reticulum (ER)- resident adaptor protein STING (stimulator of IFN genes), which activates the transcription factor interferon regulatory factor 3 (IRF3). The TANK binding kinase 1 (TBK1)/IRF3 axis results in the induction of type I IFNs, and the activation of dendritic cells (DCs) and cross-presentation of tumor antigens to activate CD8 ⁺ T cell-mediated anti-tumor immunity. STING signaling also activates the nuclear factor kappa-light-chain-enhancer of activated B cell (NF-KB) signaling axis, resulting in a pro-inflammatory response, but not in the activation of the DCs and CD8 ⁺ T-cells that are required for anti-tumor immunity.

Upon recognition of 2’3’-cGAMP, STING translocates from the endoplasmic reticulum through the Golgi apparatus, allowing the recruitment of TANK-binding kinase 1 (TBK1), and the activation of the transcription factors IRF3 and NF-KB. The carboxyl-terminal tail (C -terminal tail or CTT) region of STING is necessary and sufficient to activate TBK1 and stimulate the phosphorylation of IRF3; it also is involved in NF-KB signaling. The CTT is an unstructured stretch of approximately 40 amino acids that contains sequence motifs required for STING phosphorylation and recruitment of IRF3. IRF3 and NF-KB downstream signaling is attributed to specific sequence motifs within the C-terminal tail (CTT) of STING that are conserved among vertebrate species. Modular motifs in the CTT, which include IRF3-, TBK1-, and TRAF6-binding modules, control the strength and specificity of cell signaling and immune responses.

Depending on the species and the respective characteristics of their STING CTT discrete elements, the IRF3 and NF-KB downstream responses can be affected, and sometimes opposite. The STING CTT elements dictate and finely tune the balance between the two signaling pathways, resulting in different biological responses. In human and mouse immune cells, for example, STING-dependent IRF3 activation results predominantly in a type I interferon response. STING signaling in human cells also drives a pro-inflammatory response through canonical and possibly non-canonical NF-KB pathways via TRAF6 recruitment. Human STING residue S366 (see, e.g., SEQ ID N0s:305-309) is a primary TBK1 phosphorylation site that is part of an LxIS motif in the CTT, which is required for IRF3 binding, while a second PxPLR motif, including residue L374, is required for TBK1 binding. The LxIS and PxPLR motifs are highly conserved in all vertebrate STING alleles. In other species, STING signaling results predominantly in the activation of the NF-KB signaling axis. For example, the zebrafish CTT, which is responsible for hyperactivation of NF-KB signaling, contains an extension with a highly conserved PxExxD motif at the extreme C-terminus that is not present in human and other mammalian STING alleles; this motif shares similarity with tumor necrosis factor receptor-associated factor 6 (TRAF6) binding sites. While the role of TRAF6 in human STING signaling is non- essential, TRAF6 recruitment is essential for zebrafish STING-induced NF-KB activation. A human-zebrafish STING chimera, in which human STING was engineered to contain the zebrafish STING CTT module DPVETTDY, induced more than 100-fold activation of NF-KB activation, indicating that this region is necessary and sufficient to direct enhanced NF-KB signal activation. The addition of the zebrafish CTT also resulted in an increased STING interferon response (see, de Oliveira Mann et al. (2019) Cell Reports 27:1165-1175).

The differences among species in the balance between IRF3 and NF-KB signaling is exploited herein to produce modified STING proteins that have reduced NF-KB signaling, and/or optionally, increased IRF3 signaling, so that when the STING protein is delivered to and expressed in the TME, the resulting response is an increased anti-tumor/anti-viral response, compared to the unmodified STING protein.

In some embodiments, STING proteins from species that have low or no NF- KB signaling activity are provided in delivery vehicles, including any of the immunostimulatory bacteria described herein or known to those of skill in the art, as well as in other delivery vehicles, such as viral vectors, including oncolytic vectors, minicells, exosomes, liposomes, and in cells, such as T-cells that are used in cell therapy and used to deliver vehicles, such as bacteria and oncolytic vectors.

The non-human STING proteins can be, but are not limited to, STING proteins from the following species: Tasmanian devil (Sarcophilus harrisip SEQ ID NO:349), marmoset (Callithrix jacchus,' SEQ ID NO:359), cattle (Bos taurus, SEQ ID NO:360), cat (Fells catiis, SEQ ID NO:356), ostrich (Struthio camelus australis,' SEQ ID NO:361), crested ibis (Nipponia nippon, SEQ ID NO:362), coelacanth (Latimeria chalumnae,' SEQ ID NOs:363-364), boar (Sus scrofa, SEQ ID NO:365), bat (Rousettus aegyptiacus,' SEQ ID NO:366), manatee (Trichechus manatus latirostris,' SEQ ID NO:367), ghost shark (Callorhinchus milii, SEQ ID NO:368), and mouse (Mus musculus SEQ ID NO:369). These vertebrate STING proteins readily activate immune signaling in human cells, indicating that the molecular mechanism of STING signaling is shared in vertebrates (see, de Oliveira Mann et al. (2019) Cell Reports 27:1165-1175).

In other embodiments, the non-human STING proteins contain any of the constitutive STING activation and gain-of-function mutations, at corresponding loci in the non-human STING corresponding to those in human STING, described above (see, Example 17 below, which provides exemplary alignments and corresponding mutations in various species; see, also, Figures 1-13).

In other embodiments, chimeras of STING proteins are provided. In the chimeras, the CTT region, or portion(s) thereof that confers or participates in NF-KB signaling/activity, of a first species STING protein is replaced with the corresponding CTT or portion(s) thereof from a second species, whose STING protein has lower or very little, less than human, NF-KB signaling activity. Generally, the first species is human, and the replacing CTT or portion(s) thereof is from the STING of a species such as Tasmanian devil, marmoset, cattle, cat, ostrich, boar, bat, manatee, crested ibis, coelacanth, and ghost shark, which have much lower NF-KB activity. This thereby results in a STING protein that induces type I interferon, which is important for anti-tumor activity, and that has limited or no NF-KB activity, which is not desirable in an anti-tumor therapy. The chimeras can further include the human constitutive STING activation and gain-of-function mutations in corresponding loci to increase or render type I interferon activity constitutive. In all embodiments, the TRAF6 binding motif can be deleted to further decrease or eliminate activity that is not desirable in an anti-tumor therapeutic. These non-human STING proteins, chimeras, and mutants are provided in delivery vehicles, such as any described herein or known to those of skill in the art, including oncolytic viral vectors, cells, such as stem cells and T-cells that are used in cell therapies, exosomes, minicells, liposomes, and the immunostimulatory bacteria provided herein, which accumulate in tumorresident immune cells, and deliver encoded proteins to the tumor microenvironment and to tumors. The non-human STING proteins, modified STING proteins, and STING chimeras, are for use as therapeutics for the treatment of tumors as described herein, or for use in other methods known to those of skill in the art. Pharmaceutical compositions containing the STING proteins, delivery vehicles, and encoding nucleic acids also are provided. d. Other Gene Products that Act as Cytosolic DNA/RNA Sensors and Constitutive Variants Thereof

Other gene products that sense or interact with cytosolic nucleic acids are the retinoic acid-inducible gene I (RIG-I)-like receptors (RLRs), which include RIG-I and MDA5 (melanoma differentiation-associated protein 5). RLRs are cytoplasmic sensors of viral dsRNA and nucleic acids secreted by bacteria, and include RIG-I, MDA5, and LGP2 (laboratory of genetics and physiology 2). Upon the binding of a ligand, such as a viral dsRNA, RIG-I and MDA5 activate the mitochondrial antiviralsignaling adaptor protein, or MAVS, which recruits tumor necrosis factor (TNF) receptor-associated factors (TRAFs), to assemble a signaling complex at the outer membranes of the mitochondria. Downstream signaling components further are recruited by TRAFs, resulting in the phosphorylation and activation of IRF3 (interferon regulatory factor 3), IRF7, NF-KB (nuclear factor kappa-light-chain- enhancer of activated B cells), and AP-1 (activator protein 1). As result, the expression of IFNs, proinflammatory cytokines, and other genes involved in pathogen clearance, is induced (see, e.g., Lu and MacDougall (2017) Front. Genet. 8: 118). Like STING, the constitutive activation of MDA5 and RIG-I due to gain-of-function mutations leads to the induction of type I IFNs, which can be leveraged to enhance the anti-tumor immune response in the immunostimulatory bacteria. i. RIG-I

Retinoic acid-inducible gene I (RIG-I), also known as DDX58 (DEXD/H-box helicase 58), is another protein whose constitutive activation has been linked to the development of interferonopathies, such as atypical Smith-Magenis syndrome. RIG-I, like MDA5/IFIH1, is a member of the RIG-I-like receptor (RLR) family, and is a 925- residue cytosolic pattern recognition receptor that functions in the detection of viral dsRNA. RIG-I initiates an innate immune response to viral RNA through independent pathways that promote the expression of type I and type III IFNs and proinflammatory cytokines (see, e.g., Jang et al. (2015) Am. J. Hum. Genet. 96:266- 274; and Lu and MacDougall (2017) Front. Genet. 8:118).

Atypical Smith-Magenis syndrome, without hallmark dental anomalies, but with variable phenotypes, including glaucoma, aortic calcification and skeletal abnormalities, has been found to be caused by mutations in the DEXD/H-box helicase 58 gene (DDX58), which encodes retinoic acid-inducible gene I (RIG-I). In particular, the mutations E373A and C268F in DDX58 were identified as causing gain-of- function in RIG-I. Elevated amounts of mutated DDX58 were associated with a significant increase in the basal levels of NF-KB reporter gene activity, and this activity was further increased by stimulation with the dsRNA analog poly(I:C). The RIG-I mutations also induced IRF3 phosphorylation and dimerization at the basal level, and led to increased expression of IFNB1, interferon-stimulated gene 15 ISG15), and chemokine (C-C motif) ligand 5 (CCL5) in both basal, and poly(I:C) transfected HEK293FT cells. These results indicate that the mutated DDX58/RIG-I results in constitutive activation, leading to increased IFN activity and IFN-stimulated gene expression (see, e.g., lang et al. (2015) Am. J. Hum. Genet. 96:266-274; and Lu and MacDougall (2017) Front. Genet. 8: 118). Tumor-targeting immunostimulatory bacteria provided herein can be modified to encode RIG-I/DDX58 (see, e.g., SEQ ID NOG 11) with gain-of-function mutations such as, but not limited to, E373A and C268F, singly, and in combination. ii. MDA5/IFIH1

Another interferonopathy gene is the IFN-induced with helicase C domaincontaining protein 1 (IFIH1), also known as melanoma differentiation-associated protein 5 (MDA5), which is a member of the RIG-I-like family of cytoplasmic DExD/H box RNA receptors. MDA5, encoded by IFIH1, is a 1,025 amino acid cytoplasmic pattern-recognition receptor that senses viral double-stranded RNA (dsRNA) and secreted bacterial nucleic acids in the cytoplasm, and activates type I IFN signaling through an adaptor molecule, MAVS (mitochondrial antiviral-signaling protein). MAVS recruits tumor necrosis factor (TNF) receptor-associated factors (TRAFs), which in turn recruit downstream signaling components, resulting in the phosphorylation and activation of IRF3 (interferon regulatory factor 3), IRF7, NF-KB (nuclear factor kappa-light-chain-enhancer of activated B cells), and AP-1 (activator protein 1). This results in the expression of IFNs, proinflammatory cytokines, and other genes involved in pathogen clearance (see, e.g., Rutsch et al. (2015) Am. J. Hum. Genet. 96:275-282; Rice et al. (2014) Nat. Genet. 46(5):503-509; and Lu and MacDougall (2017) Front. Genet. 8:118).

Gain-of-function (GOF) IFIH1 variants occur in subjects with autoimmune disorders, including Aicardi-Goutieres syndrome (AGS) and Singleton-Merten syndrome (SMS), which are characterized by prominent vascular inflammation. AGS is an inflammatory disease particularly affecting the brain and skin, and is characterized by an upregulation of interferon-induced transcripts. AGS typically occurs due to mutations in any of the genes encoding DNA exonuclease TREX1, the three non-allelic components of the RNase H2 endonuclease complex, the deoxynucleoside triphosphate triphosphohydrolase SAMHD1, and the doublestranded RNA editing enzyme AD ARI . Some patients with AGS do not have mutations in any of these genes, but have GOF mutations in IFIH1, indicating that this gene also is implicated in AGS. Singleton-Merten syndrome is an autosomal- dominant disorder characterized by abnormalities in the blood vessels (e.g., calcification), teeth (e.g., early-onset periodontitis, root resorption), and bones (e.g., osteopenia, acro-osteolysis, osteoporosis). Interferon signature genes are upregulated in Singleton-Merten syndrome patients, which was linked to GOF mutations in IFIH1 (see, e.g., Rice et al. (2014) to. Genet. 46(5):503-509; and Rutsch et al. (2015) Am. J. Hum. Genet. 96:275-282).

The IFN-P reporter stimulatory activity of wild-type IFIH1, and six IFIH1 GOF mutants identified in AGS patients (R720Q, R779H, R337G, R779C, G495R, D393V), was compared in HEK293T cells, which express low levels of endogenous viral RNA receptors. Wild-type IFIH1 was induced upon binding of the long (> 1 kb) dsRNA analog polyinosinic-polycytidylic acid (poly(I:C)), but not by a short 162 bp dsRNA, and had minimal activity in the absence of exogenous RNA. The IFIH1 mutants displayed a significant induction of IFN signaling in response to the short 162 bp dsRNA, in addition to robust signaling in response to poly(I:C). The mutants also displayed a 4-10 fold higher level of baseline signaling activity in the absence of exogenous ligand (Rice et al. (2014) Nat. Genet. 46(5):503-509).

Another gain-of-function IFIH1 mutation, R822Q, was identified as causing Singleton-Merten syndrome by triggering type I IFN production, and leading to early arterial calcification, as well as dental inflammation and resorption. HEK293T cells (which have the lowest endogenous IFIH1 expression levels) were used to overexpress wild-type and R822Q MDA5. Wild-type IFIH1 expression led to an increase in the expression of IFNB 1 (interferon, beta 1, fibroblast) in a dosedependent manner, whereas the mutated IFIH1 led to approximately 20-fold more IFNB1 expression. Following stimulation with the dsRNA analog poly(I:C), R822Q IFIH1 resulted in higher levels of IFNB1 expression than wild-type IFIH1, indicating that R822Q IFIH1 is hyperactive to non-self dsRNA. There was also higher expression of interferon signature genes, such as IFI27, IFI44L, IFIT1, ISG15, RSG15, RSAD2, and SIGLEC1 in whole-blood samples from Singleton-Merten syndrome patients, which was in agreement with the higher expression level of IFNB1 by R822Q //'7T/7 (see, e.g., Rutsch etal. (2015) 4m. J. Hum. Genet. 96:275-282).

The interferon signature observed in patients with another IFIH1 GOF mutation, A489T, is indicative of a type I interferonopathy; IFIH1 A489T is associated with increased interferon production and phenotypes resembling chilblain lupus, AGS, and SMS (see, e.g., Bursztejn etal. (2015) Br. J. Dermatol. 173(6): 1505- 1513). The A489T variant not only resulted in IFN induction following stimulation with the long dsRNA analog poly(I:C), but also with short dsRNA. Two additional gain-of-function mutations in IFIH1, T331I and T331R, were identified in patients with SMS phenotypes, who presented with a significant upregulation of IFN-induced transcripts. The T331I and T331R variants resulted in increased expression of IFN- , even in the absence of exogenous dsRNA ligand, consistent with the observed constitutive activation of MDA5 (see, e.g., Lu and MacDougall (2017) Front. Genet. 8: 118).

A946T is another IFIH1 GOF mutation that leads to the increased production of type I IFN, promoting inflammation and increasing the risk of autoimmunity. The A946T mutation in IFIHI results in additive effects when combined with the TMEM173 R232 allele and the G207E GOF mutation in STING, leading to a severe early-onset phenotype with features similar to SAVI (see, e.g., Keskitalo et al. (2018) preprint, available from doi.org/10.1101/394353). G821S is a GOF mutation in IFIHI which has been shown to lead to spontaneously developed lupus-like autoimmune symptoms in a mouse model (see, e.g., Rutsch et al. (2015) Am. J. Hum. Genet. 96:275-282), while the IFIHI missense mutations A452T, R779H, and L372F, identified in individuals with AGS, were shown to cause type I interferon overproduction (see, e.g., Oda et al. (2014) Am. J. Hum. Genet. 95: 121-125).

The tumor-targeting immunostimulatory bacteria provided herein can be modified to encode MDA5/IFIH1 (see, e.g., SEQ ID NO:310) with gain-of-function mutations selected from T331I, T331R, R337G, L372F, D393V, A452T, A489T, G495R, R720Q, R779H, R779C, G821S, R822Q, and A946T, singly, or in any combination. iii. IRF7

Constitutively active forms of IRF7 (or IRF-7) include mutants in which different C-terminal serines are substituted by phosphomimetic Asp, including IRF7(S477D/S479D), IRF7(S475D/S477D/S479D), and IRF7(S475D/S476D/S477D/S479D/S483D/S487D). IRF7(S477D/S479D) is a strong transactivator for IFNA and RANTES gene expression, and stimulates gene expression, even in the absence of virus infection. IRF7(S475D/S477D/S479D), and IRF7(S475D/S476D/S477D/S479D/S483D/S487D) do not further augment the transactivation activity of IRF7(S477D/S479D), but the transactivation activity of all three mutants is stimulated further by virus infection. The mutant IRF7(A247-467), which localizes to the nucleus in uninfected cells, is a very strong constitutive form of IRF7; it activates transcription more than 1500-fold higher than wild-type IRF7 in unstimulated and virus infected cells (see, e.g., Lin etal. (2000) J. Biol. Chem. 275 (44): 34320-34327). The immunostimulatory bacteria provided herein, can encode and express constitutively active IRF7 mutants, including those with replacements at residues 475-477, 479, 483, and 487, and those with amino acid deletions. The immunostimulatory bacteria encode these proteins on plasmids under the control of promoters and, any other desired regulatory signals recognized by mammalian hosts, including humans. e. Other Type I IFN Regulatory Proteins

Other proteins involved in the recognition of DNA/RNA that activate type I IFN responses can be mutated to generate constitutive type I IFN expression. The unmodified and/or modified proteins can be encoded in the immunostimulatory bacteria provided herein, to be used to deliver the protein to the tumor microenvironment, such as to tumor-resident immune cells, to increase expression of type I IFN.

These proteins include, but are not limited to, proteins designated TRIM56, RIP1, Sec5, TRAF2, TRAF3, TRAF6, STAT1, LGP2, DDX3, DHX9 (DDX9), DDX1, DDX21, DHX15, DHX33, DHX36, DDX60, and SNRNP200.

Gain-of-function variants can be produced, such as by screening and/or by mutagenesis. Site-directed mutagenesis can be performed in vitro to identify mutations with enhanced activity, that lead to higher level and/or constitutive type I IFN expression. Intact genomic DNA can be obtained from non-related patients experiencing auto-immune and auto-inflammatory symptoms, and from healthy individuals, to screen for and identify other products whose expression leads to increased or constitutive type I IFN expression. Whole exome sequencing can be performed, and introns and exons can be analyzed, such that proteins with mutations in the pathways associated with the increased or constitutive expression of type I interferon are identified. After identification of mutations, cDNA molecules encoding the full-length gene, with and without the identified mutation(s), are transfected into a reporter cell line that measures expression of type I interferon. For example, a reporter cell line can be generated where the expression of luciferase is placed under the control of the promoter for IFN-p. A gain-of-function mutant that is constitutively active will promote the expression of IFN-P, whereas the unstimulated wild-type protein will not. Stimulation can be by virus infection, bacterial infection, bacterial nucleic acids, LPS, dsRNA, poly(I:C), or by increasing exogenous levels of the protein’s ligand (e.g., CDNs). Identified proteins also include those that enhance an immune response to an antigen(s) of interest in a subject. The immune response comprises a cellular or humoral immune response characterized by one or more of: (i) stimulating type I interferon pathway signaling; (ii) stimulating NF-KB pathway signaling; (iii) stimulating an inflammatory response; (iv) stimulating cytokine production; (v) stimulating dendritic cell development, activity, or mobilization; (vi) any other responses indicative of a product whose expression enhances an immune response; and (vii) a combination of any of (i)-(vi).

3. Antibodies and Antibody Fragments

Advances in antibody engineering have led to the creation of recombinant antibody fragments that have many improvements over conventional monoclonal antibodies, especially in terms of manufacturing, tissue penetration, and ease of use. An example of these is the single-chain fragment variable (scFv), consisting of the vanable regions of the heavy (VH) and light (VL) chains of the antibody binding site, joined together by a flexible peptide linker that is generally the (648)3 sequence (see, e.g., Weisser et al. (2009) Biotechnol. Adv. 27(4): 502-520). Other examples include scFv-Fc antibody fragments, in which the VH domain of the scFv is linked to an Fc region, as well as dimers and multimers and other combinations and forms of antibodies that are known to those of skill in the art. Antibody fragments and other constructs allow for targeting of antigens in a manner that can be encoded on a plasmid and delivered, as exemplified herein, by an immunostimulatory bacterium. Examples of antigens to target, include, tumor antigens, including those discussed in the Examples below, and other exemplary targets discussed below and elsewhere herein. Any target known to the skilled person is included, as are neo-antigens formed upon changes in conformation or structure. a. TGF-p

Transforming growth factor beta (TGF-P) is a pleiotropic cytokine with numerous roles in embryogenesis, wound healing, angiogenesis, and immune regulation. It exists in three isoforms in mammalian cells, TGF- i, TGF- 2, and TGF- 3; TGF-pl is the most predominant in immune cells (see, e.g., Esebanmen et al. (2017) Immunol. Res. 65:987-994). TGF- ’s role as an immunosuppressant is arguably its most dominant function. Its activation from a latent form in the tumor microenvironment, in particular, has profound immunosuppressive effects on DCs and their ability to tolerize antigen-specific T-cells. TGF-P also can directly convert Thl CD4 ⁺ T-cells to immunosuppressive Tregs, further promoting tumor tolerance (see, e.g., Travis etal. (2014) Amu. Rev. Immunol. 32:51-82). Based on its tumorspecific immunosuppressive functions, and irrespective of its known cancer cell growth and metastasis-promoting properties, inhibition of TGF-P is a cancer therapy target. High levels of TGF-P signaling have been demonstrated in several human tumor types, including colorectal cancer (CRC), hepatocellular carcinoma (HCC), pancreatic ductal adenocarcinoma (PDAC), and non-small-cell lung cancer (NSCLC) (see, e.g., Colak et al. (2017) Trends Cancer 3(1): 56-71). Systemic inhibition of TGF- P can lead to unacceptable autoimmune toxicities, and its inhibition should be localized to the tumor microenvironment. One way to accomplish this is to create a soluble TGF-p receptor that acts as a decoy for binding TGF-P (see, e.g., Zhang et al. (2008) J Immunol. 181:3690-3697). As such, a tumor-targeting immunostimulatory bacteria, containing a TGF-P receptor decoy, provided herein, can bind and remove TGF-P from the tumor microenvironment, thereby breaking tumor immune tolerance and stimulating anti-tumor immunity.

In addition to TGF-beta binding decoy receptors, other TGF-beta polypeptide antagonists, that can bind to and eliminate or reduce TGF-P in the tumor microenvironment, thereby breaking tumor immune tolerance and stimulating antitumor immunity, are included. Exemplary of such are anti-TGF-beta antibodies or antibody fragments, anti-TGF-beta receptor antibodies or antibody fragments, and soluble TGF-beta antagonist polypeptides.

Immunostimulatory bacteria are provided herein that accumulate in the tumor microenvironment, in tumors, and in particular, in tumor-resident immune cells, and that contain plasmids encoding TGF-beta polypeptide antagonists, including, for example, TGF-beta binding decoy receptors (TGF-P receptor decoys), anti-TGF-beta antibodies or antibody fragments, anti-TGF-beta receptor antibodies or antibody fragments, and soluble TGF-beta antagonist polypeptides. The antibody fragments can include any known in the art, or described herein, such as, but not limited to, scFvs and scFv-Fcs. b. Bispecific scFvs and T-Cell Engagers

The use of scFvs has been improved by increasing the valency of binding to the target, often through the use of one or more scFv fragments (bis-specific, tri- specific, etc.), joined together by a long linker. Bi-specific T-cell engager (available under the trademark BiTE®) constructs are a class of artificial bispecific monoclonal antibodies that are utilized in cancer immunotherapy, and are formed by linking two single-chain variable fragments (scFvs), such that one scFv binds CD3 on the surface of cytotoxic T-cells, and the other binds a specific tumor-associated antigen (TAA). BiTEs® thus target T-cells to tumor cells, stimulating T-cell activation, cytokine production and tumor cell cytotoxicity independently of MHC class I or costimulatory molecules. Two examples of bi-specific T-cell engagers (BiTEs®) have been approved by the FDA, including catumaxomab (Removab®), which is directed against the tumor antigen EpCAM, and against CD3, and is used in the treatment of malignant ascites, and blinatumomab (BLINCYTO®), a BiTE® antibody directed against CD19 and CD3, which is used for the treatment of relapsed, refractory acute lymphoblastic leukemia (ALL; see, e.g., Ahamadi-Fesharaki etal. (2019) Afo/. Ther. Oncolytics 14:38-56). Other bi-specific T-cell engagers (BiTEs®) target a variety of antigens, including, for example, carcinoembryonic antigen (CEA), prostate-specific membrane antigen (PSMA), EGFR, EphA2, HER2/neu, ADAM17/TACE, prostate stem cell antigen (PSCA), Delta-like ligand 3 (DDL3), and melanoma-associated chondroitin sulfate proteoglycan (MCSP). As exemplified herein, a BiTE® antibody also can be expressed from a plasmid following delivery by an immunostimulatory bacterium. c. Anti-PD-1, Anti-PD-Ll and Anti-CTLA-4 Antibodies i. Anti-PD-l/Anti-PD-Ll Antibodies

Programmed cell death protein 1 (PD-1) is an immune-inhibitory receptor that is involved in the negative regulation of immune responses. Its cognate ligand, programmed death-ligand 1 (PD-L1), is expressed on antigen-presenting cells (APCs), and upon binding to PD-1 on T-cells, leads to loss of CD8 ⁺ T-cell effector function, inducing T-cell tolerance. The expression of PD-L1 is often associated with tumor aggressiveness and reduced survival in certain human cancers (see, e.g., Gao et al. (2009) Clin. Cancer Res. 15(3): 971-979).

Antibodies designed to block immune checkpoints, such as anti -PD-1 (for example, pembrolizumab, and nivolumab) and anti-PD-Ll (for example, atezolizumab, avelumab, and durvalumab) antibodies can prevent T-cell anergy and break immune tolerance. Only a fraction of treated patients, however, exhibit clinical benefit, and those that do, often present with autoimmune-related toxi cities (see, e.g., Ribas (2015) A. Engl. J. Med. 373(16):1490-1492; and Topalian et al. (2012) A Engl. J. Med. 366(26):2443-2454). Besides acquiring toxicity, anti-PD-l/anti-PD-Ll therapy often leads to resistance, and the concomitant use of anti-CTLA-4 antibodies (for example, ipilimumab) has shown limited success in clinical trials, with significantly additive toxicity. To limit the toxicity and enhance the potency of PD- 1/PD-L1 blockade, an immunostimulatory bacterium, containing a plasmid encoding an antibody or antibody fragment, such as an scFv and scFv-Fc, and others known in the art and/or described herein, against PD-1 or against PD-L1, will synergize with activation of immune cells to potentiate anti-tumor immunity. ii. Anti-CTLA-4 Antibodies

CTLA-4 (cytotoxic T-lymphocyte-associated protein 4), also known as CD152 (cluster of differentiation 152), is another immune-inhibitory receptor that functions as an immune checkpoint, and downregulates immune responses. CTLA-4 is constitutively expressed in regulatory T-cells (Tregs, or Tregs), and contributes to their inhibitory function, but is upregulated in conventional T-cells only after activation. CTLA-4 functions as an immune checkpoint by transmitting inhibitory signals to T- cells. CTLA-4 is homologous to the T-cell co-stimulatory protein, CD28, and both molecules bind to CD80 (also known as B7-1 or B7.1) and CD86 (also known as B7- 2 or B7.2) ligands on antigen-presenting cells (APCs). The binding of CTLA-4 to the ligands transmits an inhibitory signal to T-cells, whereas the binding of CD28 transmits a stimulatory signal.

Following T-cell activation, CTLA-4 receptors are induced, which then outcompete CD28 receptors on T-cells, for binding to CD80 and CD86 ligands on the surfaces of APCs. CTLA-4 binds to CD80 and CD86 with greater affinity and avidity than CD28, thus enabling it to outcompete CD28 for its ligands, resulting in the transmittal of inhibitory signals to T-cells, and an immune inhibitory response. T-cell activation through the T-cell receptor and CD28 leads to increased expression of CTLA-4.

Optimal T-cell priming requires co-stimulatory signals resulting from the ligation of T-cell CD28 with CD80 and/or CD86. The blockade of CTLA-4 from binding to these ligands thus enhances T-cell priming, and allows for the induction of an anti-tumor immune response.

In some embodiments, the immunostimulatory bacterial strains provided herein contain plasmids encoding anti-CTLA-4 antibodies, including fragments thereof, such as, but not limited to, anti-CTLA-4 scFvs (see, e.g., copending PCT application International PCT Application No. PCT/US2020/060307 for sequences of exemplary human anti-CTLA-4 scFv and scFv-Fc fragments; see, also, Example 25). d. Additional Exemplary Checkpoint Targets

Exemplary immune checkpoint targets for which an scFv, or any other recombinant antibody fragment against them can be prepared, or are exemplified herein include, but are not limited to, those listed in the table below:

4. Combinations of Immunomodulatory Proteins can have Synergistic Effects and/or Complementary Effects

Cytokines are powerful modulators of the anti-tumor immune response. Cytokine combinations are known to have profound synergistic effects on different immune compartments involving T-cells, NK cells, and myeloid cells (including dendritic cells and macrophages). Cytokines are known to play major roles in antigen priming by dendritic cells, survival and proliferation of innate immune cells and antigen-specific T-cells, and the cytotoxic activity of NK and T-cells. Cytokine combinations must be properly chosen to maximize biological responses and enhance anti-tumor immunity. For example, in a murine model of hepatitis, IFN-a alone was found to enhance the CD8 ⁺ T-cell cytolytic function of virally infected cells, while IL- 15 alone enhanced the proliferation of activated lymphocytes. Together, they maximally suppressed hepatitis B (HBV) infection (see, e.g., Di Scala et al. (2016) J. Virol. 90(19): 8563-8574). In another example, combinations of the cytokines IL-15 + IL- 18, and IL- 15 + IL-21, were able to enhance the production of IFN-y from human NK and T-cells (see, e.g., Strengell et al. (2003) J. Immunol. 170(11):5464-5469). In another example, IL-2 + IL- 18 synergized to enhance IFN-y production and increase the cytolytic function of CD4 ⁺ T-cells, CD8 ⁺ T cells, and NK lymphocytes (see, e.g., Son etal. (2001) Cancer Res. 61(3):884-888). Additionally, IL-12 and IL-18 were found to synergize to promote antigen-CD3 T-cell ligation-independent production of IFN-y from human T-cells (see, e.g., Tominaga et al. (2000) Int. Immunol. 12(2): 151- 160). Combinations of cytokines are powerful enhancers of T-cell function, but the FDA-approved anti-cancer cytokines are too toxic to be dosed systemically and are thus rarely used, and combinations of systemically-administered cytokines only compound the toxicity (see, e.g. , Conlon et al. (2019) J. Interferon Cytokine Res. 39(1):6-21).

The immunostimulatory bacteria provided herein solve these problems. Provided are immunostimulatory bacteria containing plasmids encoding multiple therapeutic products, such as immunomodulatory proteins, that allow for tumorspecific delivery of cytokine combinations and/or combinations with other therapeutic products, such as the inducers of type I interferon discussed herein, and others, including co-stimulatory molecules, chemokines, and antibodies and fragments thereof. These immunostimulatory bacteria achieve powerful and synergistic immuno-activation without the systemic toxicities and pharmacokinetic (PK) liabilities associated with direct IV administration of the cytokines and other therapeutic products.

Combinations of therapeutic products that can be encoded on the plasmids in the immunostimulatory bacteria provided herein include, but are not limited to, for example, two or more cytokines; one or more cytokines and an inducer of type I IFN (e.g., STING, IRF3, IRF7, MDA5, RIG-I, and constitutively active, GOF variants thereof), and/or a co-stimulatory molecule (e.g., 4-1BBL and 4-lBBLAcyt); a TGF-P decoy receptor and one or more cytokines; a TGF-P decoy receptor and an inducer of type I IFN; a TGF-P decoy receptor, one or more cytokines, and/or an inducer of type I IFN, and/or a co-stimulatory molecule; an antibody (e.g., against an immune checkpoint, such as CTLA-4) and one or more cytokines; an antibody and an inducer of type I IFN; an antibody, one or more cytokines, and/or an inducer of type I IFN, and/or a co-stimulatory molecule; a co-stimulatory molecule agonist (e.g., a CD40 agonist) and one or more cytokines; a co-stimulatory molecule agonist and an inducer of type I IFN; and a co-stimulatory molecule agonist, one or more cytokines, and/or an inducer of type I IFN, and/or a co-stimulatory molecule.

As discussed below, the multiple therapeutic product expression cassettes can include single promoter constructs and/or dual/multiple promoter constructs, as well as post-transcriptional regulatory elements, and other regulatory elements, such as enhancers, polyadenylation signals, terminators, signal peptides, etc. The nucleic acid sequences can be codon optimized to increase protein expression, and generally, are under control of a eukaryotic promoter. Particular constructs and details thereof are described elsewhere herein.

Among the immunostimulatory bacteria provided herein are those that contain plasmids encoding immunostimulatory proteins e.g., cytokines, chemokines, costimulatory molecules), and/or gene products with gain-of-function mutations that increase immune responses in the tumor microenvironment (e.g., cytosolic DNA/RNA sensors that induce type I IFN), and/or antibodies and fragments thereof, and/or other therapeutic products that enhance the anti-tumor response, such as TGF-P and/or IL-6 decoy receptors, and/or TGF-P antagonizing polypeptides. These immunostimulatory bacteria that encode the cytokines, gain-of-function products/type I IFN pathway proteins, and/or chemokines, and/or co-stimulatory molecules, and/or antibodies and fragments thereof, such as single-chain antibodies, and other therapeutic products discussed herein, include the immunostimulatory bacteria that preferentially infiltrate the tumor microenvironment, tumors, and tumor-resident immune cells. The immunostimulatory bacteria also include those in which the genome is modified so that they induce less cell death in tumor-resident immune cells, whereby the immunostimulatory bacteria accumulate in tumor-resident myeloid cells, to achieve high level ectopic expression of multiplexed genetic payloads in the target cells, and deliver the therapeutic products/immunomodulatory proteins to the tumor microenvironment (TME), to stimulate the immune response against the tumor. In particular, the immunostimulatory bacteria provided herein include up to about 8 or 8 modifications as described herein, including, but not limited to, adenosine auxotrophy, csgD~,pagP~, msbR, flagellin" (such as, for example, fliC7fljB~'), purl" , ansB', asd, and any other modifications described herein or known to improve targeting to, or accumulation in, the tumor microenvironment and/or tumor-resident myeloid cells, or to improve safety and tolerability (allowing for a higher dose), reduce the immunosuppressive cytokine profile, improve T-cell quality and function, limit replication in healthy tissues, eliminate biofilms, and improve the anti-tumor immune response, or to impart any of the desirable and advantageous properties discussed elsewhere herein.

The immunostimulatory bacteria further can encode other therapeutic products, such as a tumor antigen from the subject’s tumor, to enhance the response against the particular tumor. Any of the immunostimulatory bacteria provided herein and described above and below can be modified to encode the therapeutic products, such as cytokines, chemokines, co-stimulatory molecules and gain-of-function type I IFN pathway product(s). The therapeutic products are encoded on a plasmid under control of a promoter recognized by the host, and any other desired regulatory sequences recognized in a eukaryotic, such as a human, or other animal, or mammalian, subject. Generally, the nucleic acid encoding the product is under the control of an RNA polymerase II promoter. Additionally, any of the bacteria described herein for modification, such as any of the strains of Salmonella, Shigella, E. coli, Bifidobacteriae , Rickettsia, Vibrio, Listeria, Klebsiella, Bordetella, Neisseria, Aeromonas, Francisella, Cholera, Corynebacterium, Citrobacter , Chlamydia, Haemophilus, Brucella, Mycobacterium, Mycoplasma, Legionella, Rhodococcus, Pseudomonas, Helicobacter, Bacillus, and Erysipelothrix, or an attenuated strain thereof, or a modified strain thereof, can be modified by introducing a plasmid containing, or by encoding on a plasmid in the bacteria, nucleic acid encoding the therapeutic product(s) under control of an RNA polymerase promoter recognized by the host. The therapeutic products are expressed in the infected subject’s cells. The immunostimulatory bacteria include those that are modified, as described herein, to accumulate in, or to preferentially infect, tumors, the TME and/or tumor-resident myeloid cells. For example, immunostimulatory bacteria that encode gain-of-function products leading to the expression of, or the constitutive expression of, type I interferon (IFN), such as IFN-beta, and/or other therapeutic products as discussed herein, further are modified to have reduced ability or no ability to infect epithelial cells, but are able to infect phagocytic cells, including tumor-resident immune cells, and/or the immunostimulatory bacteria are modified so that they do not kill the infected phagocytic cells.

As described herein, genes involved in the SPI-1 pathway, and flagella, activate the inflammasome in phagocytic cells (immune cells), triggering pyroptosis. Knocking out SPI-1 genes and genes that encode flagella, decreases or eliminates pyroptosis of phagocytic cells, and also, eliminates infection of epithelial cells, resulting in increased infection of phagocytic cells. Provided are immunostimulatory bacteria that accumulate in phagocytic cells, particularly tumor-resident immune cells, such as, for example, myeloid-derived suppresser cells (MDSCs), tumor-associated macrophages (TAMs), and dendritic cells (DCs), in which they express the genetic payloads/therapeutic products encoded on plasmids that are controlled by eukaryotic promoters, such as those recognized by RNA polymerase II, and include other eukaryotic regulatory signals, as discussed herein. Expressed therapeutic products include those that evoke immune responses, such as through pathways that increase or induce type I interferons, which increase the host response in the tumor microenvironment. The immunostimulatory bacteria also can encode immunostimulatory proteins, such as IL-2 and/or other cytokines, and/or other immunostimulatory proteins and therapeutic products, as discussed herein, further enhancing the immune response in the tumor microenvironment.

The immunostimulatory bacteria can encode products, referred to as cytosolic DNA/RNA sensors, that evoke immune responses when exposed to nucleic acids, such as RNA, DNA, nucleotides, dinucleotides, cyclic nucleotides, cyclic dinucleotides, and other such molecules, in the cytosol of cells. The immunostimulatory bacteria herein, encode modified therapeutic products that constitutively evoke immune responses, and do not require the presence of the DNA/RNA in the cytosol. Exemplary of such are components of pathways that induce type I interferon expression. The therapeutic products contemplated herein include modified forms of these cytosolic DNA/RNA sensors, that have constitutive activity or increased activity (i.e., gain-of-function products), such that type I interferon(s) is/are expressed or produced in the absence of nucleotides, dinucleotides, cyclic nucleotides, cyclic dinucleotides, and other such ligands, in the cytosol of cells. Expression of these modified products in cells, particularly in tumor cells, including tumor-resident immune cells, leads to constitutive expression of type I interferons, including interferon- , in the tumor microenvironment. Because the immunostimulatory bacteria that express these gain-of-function products accumulate in or preferentially infect tumor cells/the TME/tumor-resident immune cells, the therapeutic products are expressed in the tumor microenvironment, resulting in increased immune responses in the tumor microenvironment.

Exemplary gene products that can be encoded in the immunostimulatory bacteria and other vehicles, include, but are not limited to, proteins that sense or are involved in innate pathways that recognize cytosolic DNA/RNA and activate type I interferon production. Proteins involved in innate DNA/RNA recognition that activate type I interferon include, but are not limited to: STING, RIG-I, MDA5, IRF3, IRF7, TRIM56, RIP1/RIPK1, Sec5/EXOC2, TRAF2, TRAF3, TRAF6, STAT1, LGP2/DHX58, DDX3/DDX3X, DHX9/DDX9, DDX1, DDX21, DHX15/DDX15, DHX33/DDX33, DHX36/DDX36, DDX60, and SNRNP200. Gain-of-function mutations in any of these proteins that result in constitutive type I interferon expression are known, or can be identified, and the mutants can be delivered by the immunostimulatory bacteria to the tumor microenvironment, such as by infection of phagocytic cells, or by targeting and binding to tumor cells.

The gain-of-function mutations include those identified from individuals with disorders resulting from constitutive type I interferon expression. Exemplary of gain- of-function products are those that occur in subjects with interferonopathies. As noted above, mutations can be identified by screening, to generate gain-of-function products as well.

The nucleic acids encoding the therapeutic products further can be modified to improve properties for expression. Modifications include, for example, codon optimization to increase transcriptional efficiency in a mammalian, particularly human, subject, such as reduction of GC content or CpG dinucleotide content, removal of cryptic splicing sites, adding or removing (generally removing) CpG islands, replacement of the Shine-Dalgamo (SD) sequence, and replacement of TATA box and/or terminal signals to increase transcriptional efficiency. Codons can be optimized for increasing translation efficiency by altering codon usage bias, decreasing GC content, decreasing mRNA secondary structure, removing premature poly(A) sites, removing RNA instability motifs (ARE), reducing stable free energy of mRNA, modifying internal chi sites and ribosomal binding sites, and reducing RNA secondary structures. Additional modifications to improve expression, and to maintain or enhance bacterial fitness have been incorporated into the immunostimulatory bacteria. These are described in sections below, and detailed and exemplified in the working Examples below.

As described above, type I interferon induction pathways, mediated by host recognition of cytosolic nucleic acids, such as single-stranded and double-stranded RNA, cyclic di-nucleotides (CDNs), and other such forms of nucleic acids, induce type I IFN. There also are Toll-Like Receptor (TLR)-independent type I IFN pathways, mediated by host recognition of single-stranded (ss) and double- stranded (ds) RNA in the cytosol. These are sensed by RNA helicases, including retinoic acidinducible gene I (RIG-I), melanoma differentiation-associated gene 5 (MDA5), and through IFN-P promoter stimulator 1 (IPS- 1 ) adaptor protein-mediated phosphorylation of the IRF3 transcription factor, leading to induction of IFN- (see, e.g., Ireton and Gale (2011) Viruses 3(6):906-919). As discussed herein, proteins in these pathways can be modified, or can exist as variants, that result in constitutive expression of type I interferons (also referred to as interferon type 1), which include IFN-a and IFN-P Exemplary of such proteins are the modified STING proteins described, including those provided, herein. These include modified STING proteins with mutations that result in constitutive expression of the type I interferons so that the interferons are expressed in the absence of induction, including modified nonhuman STING proteins and human STING proteins, and also, chimeric STING proteins, such as those in which the C-terminal tail (CTT) portion is replaced with a CTT portion from a STING protein from a second species, wherein the STING protein of the second species has lower NF -KB signaling activity than the NF-KB signaling activity of human STING, and the TRAF6 binding site in the CTT optionally is deleted, and chimeric STING proteins that further include one or more gain-of-function mutations.

Therapy with the immunostimulatory bacteria provided herein can be combined with any other anti-cancer therapy, including checkpoint inhibitor therapies and, as discussed above and elsewhere herein, other cancer treatments and chemotherapy.

5. Molecules that Activate Prodrugs

The plasmids in the immunostimulatory bacteria provided herein can include nucleic acids that encode molecules, such as enzymes, that activate, such as by cleavage of a portion of, therapeutic products, such as prodrugs, including chemotherapeutic prodrugs, particularly toxins, that are activated by enzymatic cleavage. As a result, the inactive prodrug can be administered systemically, and is inactive. The plasmid-encoded activating molecule, such as an enzyme, is expressed in the tumor microenvironment after delivery of the immunostimulatory bacteria provided herein, so that the inactive prodrug is activated in the tumor microenvironment, where it exerts its anti-tumor effect. There are many examples of such prodrugs, including certain nucleosides, and toxin conjugates. Many such prodrugs and enzymes are known (see, e. ., Malekshah et al., (2016) Curr. Pharmacol. Rep. 2(6):299-308). These include prodrugs of 5 -fluorouracil (5-FU), oxazaphosphorines, platinum drugs, and enzymes, such as deaminases, nitroreductases, phosphorylases, and cytochrome P450 enzymes, and many others.

6. Immunostimulatory Bacteria that Deliver Combination Therapies The immunostimulatory bacteria herein can be used to provide more than one therapeutic product, particularly those that are for anti-cancer therapy. In general, the products are complementary products to enhance and re-program the anti-tumor immune response. The immunostimulatory bacteria, by virtue of the genomic modifications described herein, particularly the combination of several or of all of asd, flagellin" (such as fliC'lfljB')', pagP', csgD' , purl' , adenosine auxotrophy, msbB', cmsB', and any other modifications described elsewhere herein or known to those of skill in the art, accumulate in the tumor microenvironment (TME) and infect tumorresident immune cells (myeloid cells). The immunostimulatory bacteria contain plasmids, encoding complementary therapeutic products, under control of a promoter or promoters recognized by the host, and any other desired regulatory sequences recognized in a eukaryotic, such as a human, or other animal, or mammalian, subject, to effect expression of the encoded products, and, also, secretion of the products. The immunostimulatory bacteria accumulate in the TME, particularly in tumor-resident immune cells, including myeloid-derived suppressor cells (MDSCs), tumor-associated macrophages (TAMs), and dendritic cells (DCs), where the encoded therapeutic products are expressed and then secreted into the tumor microenvironment to achieve an anti-tumor effect. By appropriate combination of products, the anti-tumor effect can be enhanced by virtue of interactions of the various products with the host immune system.

As discussed elsewhere herein, the immunostimulatory bacteria, containing plasmids encoding therapeutic products, with a single promoter and open reading frame (ORF), can express two (or more) proteins through the use of viral internal ribosomal entry sites (IRES), which are cap-independent, or through translational read-through of 2A peptides (e.g, T2A, P2A, E2A, or F2A), and subsequent selfcleavage into equally expressed co-proteins. Alternatively, the genetic payloads/ therapeutic products can be expressed using dual or multiple promoter constructs, where each protein is expressed under the control of a separate promoter. A combination of single and dual/multiple promoter constructs, to express three or more proteins, also can be included on the plasmids. Generally, the nucleic acids encoding the therapeutic products are under the control of RNA polymerase II promoters. For example, promoters include, but are not limited to EF-la, CMV, SV40, UBC, CBA, PGK, GUSB, GAPDH, EIF41A, CAG, CD68, and synthetic MND promoters. The plasmids can contain other regulatory elements, such as post-transcriptional regulatory elements (PREs; e.g., WPRE, HPRE), polyadenylation signal sequences, terminators, enhancers, secretion signals (also known as signal peptides/sequences, leader peptides/sequences), DNA nuclear targeting sequences (DTS), and other regulatory elements, described elsewhere herein or known to those of skill in the art that can enhance or increase the expression and/or secretion of the encoded therapeutic products.

The genetic payloads, or therapeutic products, encoded on the plasmids include immunostimulatory proteins, such as cytokines, chemokines, and costimulatory molecules; cytosolic DNA/RNA sensors that induce type I IFN and gain- of-function/constitutively active mutants/variants thereof; antibodies and fragments thereof; bi-specific T-cell engagers (BiTEs®); soluble TGF-P receptors that act as decoys for binding TGF-P, or TGF-P antagonizing polypeptides; IL-6 binding decoy receptors; interfering RNAs (e.g., siRNA, shRNA, miRNA); and other therapeutic products as discussed below and elsewhere herein, and as known in the art; and complementary combinations of all of the preceding therapeutic products. In some embodiments, the cytokines can be encoded on the plasmid within the immunostimulatory bacteria, with a membrane anchoring motif, such as a transmembrane domain, and a collagen-binding domain.

The immunostimulatory proteins, including cytokines, chemokines, and costimulatory molecules, that can be encoded on the plasmids include, but are not limited to, IL-2, IL-7, IL-12p70 (IL-12p40 + IL-12p35), IL-15, IL- 15/IL-15Ra chain complex, IL-18, IL-21, IL-23, IL-36 gamma, interferon-a, interferon-P, IL-2 that has attenuated binding to IL-2Ra, IL-2 that is modified so that it does not bind to IL-2Ra, CXCL9, CXCL10, CXCL11, CCL3, CCL4, CCL5, proteins that are involved in or that effect or potentiate the recruitment and/or persistence of T-cells, CD40, CD40 ligand (CD40L), 0X40, 0X40 ligand (OX40L), 4-1BB, 4-1BB Ligand (4-IBBL), 4- 1BBL with a deletion in the cytoplasmic domain (4-lBBLAcyt), ICOS, CD27, members of the B7-CD28 family, and members of the tumor necrosis factor receptor (TNFR) superfamily. The immunostimulatory proteins also include truncated costimulatory molecules, such as, for example, 4-IBBL, CD80, CD86, CD27L, B7RP1 and OX40L, with a cytoplasmic domain deletion, for expression on an antigen- presenting cell (APC), where the truncated gene product is capable of constitutive immunostimulatory signaling to a T-cell through co-stimulatory receptor engagement, and is unable to counter-regulatory signal to the APC due to a deleted cytoplasmic domain.

The cytosolic DNA/RNA sensors, that induce or activate type I IFN production include, but are not limited to, STING, RIG-I, MDA5, IRF3, IRF5, and IRF7, and gain-of-function (GOF) or constitutively active variants thereof. Other proteins involved in the recognition of DNA/RNA that activate type I IFN responses, that can be mutated to generate constitutive type I IFN expression and can be encoded on the plasmids, include, but are not limited to, TRIM56, RIP1, Sec5, TRAF2, TRAF3, TRAF6, STAT1, LGP2, DDX3, DHX9, DDX1, DDX21, DHX15, DHX33, DHX36, DDX60, and SNRNP200.

Other therapeutic products that can be encoded on the plasmids delivered by the immunostimulatory bacteria herein, or that can be co-administered with the bacteria, that enhance or increase the anti-tumor response, include, but are not limited to, antibodies and fragments thereof, for example, TGF-p inhibitory antibodies; anti- IL-6 antibodies; antibodies against checkpoint inhibitors, such as PD-1, PD-L1, and CTLA-4; and antibodies against, or inhibitors of, VEGF, CD73, CD38, Siglec-15, EGFR, Her2, Mesothelin, and BCMA. Also contemplated for expression on the plasmid, or for co-administration with the immunostimulatory bacteria herein, are bispecific T-cell engagers (BiTEs®), IL-6 binding decoy receptors, TGF-beta binding decoy receptors, and TGF-beta polypeptide antagonists. Any of these antibodies, inhibitors or decoy receptors can be co-administered with the immunostimulatory bacteria herein. In some embodiments, PARP (poly (ADP)-ribose polymerase) inhibitors, histone deacetylase (HDAC) inhibitors, and/or chemotherapy, also can be co-administered with any of the therapeutic products listed above, alone, or in any combination.

Exemplary of complementary combinations of therapeutic products that can be encoded on the plasmids in the immunostimulatory bacteria herein include, but are not limited to:

IL-2 and IL-12p70; IL-2 and IL-21; IL-2, IL-12p70, and a STING GOF variant; IL-2, IL-21, and a STING GOF variant; IL-2, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); and IL-2, IL-21, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt);

IL-15/IL-15Ra and a STING GOF variant; IL-15/IL-15Ra, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); IL-15/IL-15Ra and IL-12p70; IL- 15/IL-15Ra, and IL-21; IL-15/IL-15Ra, IL-12p70, and a STING GOF variant; IL- 15/IL-15Ra, IL-21, and a STING GOF variant; IL-15/IL-15Ra, IL-12p70, a STING GOF variant, and 4-1 BBL (including 4-lBBLAcyt); and IL-15/IL-15Ra, IL-21, a STING GOF variant, and 4-1 BBL (including 4-lBBLAcyt);

IL-12p70 and IL-21; IL-12p70, IL-21 and a STING GOF variant; IL-12p70, IL-21, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); IL-12p70 and a STING GOF variant; IL-12p70, a STING GOF variant, and 4-1BBL (including 4- IBBLAcyt); IL-12p70 and IL-18; IL-12p70, IL-18 and a STING GOF variant; and IL- 12p70, IL-18, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a TGF-P decoy receptor, IL-2, and IL-12p70; a TGF-P decoy receptor, IL-2, and IL-21; a TGF-P decoy receptor, IL-2, IL-12p70, and a STING GOF variant; a TGF-P decoy receptor, IL-2, IL-21, and a STING GOF variant; a TGF-P decoy receptor, IL-2, IL-12p70, a STING GOF variant, and 4-1BBL (including 4- IBBLAcyt); and a TGF-p decoy receptor, IL-2, IL-21, a STING GOF variant, and 4- 1BBL (including 4-lBBLAcyt); a TGF-P decoy receptor, IL-15/IL-15Ra, and a STING GOF variant; a TGF-P decoy receptor, IL-15/IL-15Ra, a STING GOF variant, and 4-1 BBL (including 4- IBBLAcyt); a TGF-P decoy receptor, IL-15/IL-15Ra, and IL-12p70; a TGF-P decoy receptor, IL-15/IL-15Ra, and IL-21; a TGF-P decoy receptor, IL-15/IL-15Ra, IL- 12p70, and a STING GOF variant; a TGF-p decoy receptor, IL-15/IL-15Ra, IL-21, and a STING GOF variant; a TGF-P decoy receptor, IL-15/IL-15Ra, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); and a TGF-P decoy receptor, IL-15/IL-15Ra, IL-21, a STING GOF variant, and 4-1BBL (including 4- IBBLAcyt); a TGF-P decoy receptor, IL-12p70, and IL-21; a TGF-P decoy receptor, IL- I2p70, IL-21, and a STING GOF variant; a TGF-P decoy receptor, IL-I2p70, IL-21, a STING GOF variant, and 4-1 BBL (including 4-lBBLAcyt); a TGF-P decoy receptor and IL-12p70; a TGF-P decoy receptor, IL-12p70, and a STING GOF variant; a TGF- P decoy receptor, IL-I2p70, a STING GOF variant, and 4-1BBL (including 4- IBBLAcyt); a TGF-P decoy receptor, IL-12p70, and IL-18; a TGF-p decoy receptor, IL-12p70, IL-18, and a STING GOF variant; a TGF-P decoy receptor, IL-12p70, IL- 18, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); and a TGF-P decoy receptor and a STING GOF variant; an anti-CTLA-4 antibody, IL-2, and IL-12p70; an anti-CTLA-4 antibody, IL- 2, and IL-21; an anti-CTLA-4 antibody, IL-2, IL-12p70, and a STING GOF variant; an anti-CTLA-4 antibody, IL-2, IL-21, and a STING GOF variant; an anti-CTLA-4 antibody, IL-2, IL-12p70, a STING GOF variant, and 4-1BBL (including 4- IBBLAcyt); and an anti-CTLA-4 antibody, IL-2, IL-21, a STING GOF variant, and 4- 1BBL (including 4-lBBLAcyt); an anti-CTLA-4 antibody, IL-15/IL-15Ra, and a STING GOF variant; an anti- CTLA-4 antibody, IL-15/IL-15Ra, a STING GOF variant, and 4-1BBL (including 4- IBBLAcyt); an anti-CTLA-4 antibody, IL-15/IL-15Ra, and IL-12p70; an anti-CTLA- 4 antibody, IL-15/IL-15Ra, and IL-21; an anti-CTLA-4 antibody, IL-15/IL-15Ra, IL- 12p70, and a STING GOF variant; an anti-CTLA-4 antibody, IL-15/IL-15Ra, IL-21, and a STING GOF variant; an anti-CTLA-4 antibody, IL-15/IL-15Ra, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); and an anti-CTLA-4 antibody, IL-15/IL-15Ra, IL-21, a STING GOF variant, and 4-1BBL (including 4- IBBLAcyt); an anti-CTLA-4 antibody, IL-12p70, and IL-21; an anti-CTLA-4 antibody, IL- 12p70, IL-21, and a STING GOF variant; an anti-CTLA-4 antibody, IL-12p70, IL-21, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); an anti-CTLA-4 antibody and IL-12p70; an anti-CTLA-4 antibody, IL-12p70, and a STING GOF variant; an anti-CTLA-4 antibody, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); an anti-CTLA-4 antibody, IL-12p70, and IL-18; an anti- CTLA-4 antibody, IL-12p70, IL-18, and a STING GOF variant; an anti-CTLA-4 antibody, IL-12p70, IL-18, a STING GOF variant, and 4-1BBL (including 4- IBBLAcyt); and an anti-CTLA-4 antibody and a STING GOF variant; a CD40 agonist, IL-2, and IL-12p70; a CD40 agonist, IL-2, and IL-21; a CD40 agonist, IL-2, IL-12p70, and a STING GOF variant; a CD40 agonist, IL-2, IL-21, and a STING GOF variant; a CD40 agonist, IL-2, IL-12p70, a STING GOF variant, and 4-1 BBL (including 4-lBBLAcyt); and a CD40 agonist, IL-2, IL-21, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a CD40 agonist, IL-15/IL-15Ra, and a STING GOF variant; a CD40 agonist, IL-15/IL-15Ra, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a CD40 agonist, IL-15/IL-15Ra, and IL-12p70; a CD40 agonist, IL- 15/IL- 15Ra, and IL-21; a CD40 agonist, IL-15/IL-15Ra, IL-12p70, and a STING GOF variant; a CD40 agonist, IL-15/IL-15Ra, IL-21, and a STING GOF variant; a CD40 agonist, IL-15/IL-15Ra, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); and a CD40 agonist, IL-15/IL-15Ra, IL-21, a STING GOF variant, and 4-1BBL (including 4- IBBLAcyt); and a CD40 agonist, IL-12p70, and IL-21; a CD40 agonist, IL-12p70, IL-21, and a STING GOF variant; a CD40 agonist, IL-12p70, IL-21, a STING GOF variant, and 4- 1BBL (including 4-lBBLAcyt); a CD40 agonist and IL-12p70; a CD40 agonist, IL- 12p70, and a STING GOF variant; a CD40 agonist, IL-12p70, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); a CD40 agonist, IL-12p70, and IL-18; a CD40 agonist, IL-12p70, IL-18, and a STING GOF variant; a CD40 agonist, IL-12p70, IL- 18, a STING GOF variant, and 4-1BBL (including 4-lBBLAcyt); and a CD40 agonist and a STING GOF variant.

The following table lists exemplary products that can be encoded in plasmids in the immunostimulatory bacteria, and some effects/ character! sties of such products, including synergistic effects of the combination of the IL- 15 or IL- 15 complex, discussed and demonstrated in the Examples.

*DC = Dendritic Cells

In any of the complementary combinations above, a TGF-P decoy receptor can be replaced with a TGF-P antagonizing polypeptide. As discussed above, TGF-P decoy receptors are any that act as decoys for binding TGF-P to remove it, or are TGF-P antagonizing polypeptides (e.g., anti-TGF-beta antibodies or antibody fragments, and anti-TGF-beta receptor antibodies or antibody fragments). The STING protein or other DNA/RNA sensor that induces or activates type I IFN production, can be a GOF/constitutively active variant, or can be the wild-type protein, including the modified STING polypeptides and chimeric STING polypeptides described and provided herein. Any of the complementary combinations above also can be administered in combination with any one or more of: an anti-PD-1 antibody, an anti- CTLA-4 antibody, an anti-PD-Ll antibody, an anti-IL-6 antibody, an anti-Siglec-15 antibody, an anti-VEGF antibody, an anti-CD73 antibody, an anti-CD38 antibody, an anti-EGFR antibody, an anti-Her2 antibody, an anti-Mesothelin antibody, an anti- BCMA antibody, and antibody fragments thereof, as well as PARP inhibitors, HD AC inhibitors, or chemotherapy, and combinations thereof.

Plasmids and immunostimulatory bacteria provided herein encode combinations of therapeutic payloads. These include combinations of nucleic acid encoding any or all of the products listed in the table above. Combinations of complementary payloads were assessed, and exemplary combinations and their effects are described in the Examples. The effects of various combinations of payloads on the activation of antigen-specific T-cells, and on the secretion of CXCL10 by myeloid cells, a key chemokine involved in the recruitment of anti-tumor T-cells, were assessed. For example, combinations of the payloads can induce strong secretion of CXCL10 by bone marrow dendritic cells (BMDCs). Combining IL-36y with IL- 12p70 and STING R284G tazCTT led to higher secretion of CXCL10 and IFN-y by BMDCs. Many of the combinations induce the activation of CD8 ⁺ T-cell responses (e.g., 4-1BB expression), and the secretion of TFN-y. Particular cytokine combinations can activate T-cells. For example, the combinations of IL-12p70 + IL-15; IL-12p70 + IL-15 + IFN-a2; IL-12p70 + IL-15 + anti-4-lBB agonistic antibody; IL-12p70 + IL- 15 + IL-36y; IL-12p70 + IL-15 + IL-21; IL-12p70 + IL-21 + IL-36y; IL-12p70 + IL- 367 ⁺ IFN-a2; IL-12p70 + ZL-36y + anti-4-lBB agonistic antibody; IL-15 + ZL-36y + IFN-a2; and IL-15 + IL-36y + anti-4-lBB agonistic antibody, result in the secretion of high levels of IFN-y, but relatively low levels of IL-6, from T-cells, making them ideal combinations for optimal T-cell activation, for the induction of anti-tumor immunity in the tumor microenvironment.

Additionally, several combinations of cytokines (IL-15/IL-15R alpha chain complex, IL-12, IL-12p70, IL-15, IL-21, and IL-36y) and 4-1BB engagement, activate T-cells to secrete high levels of IFN-y, with and without TCR stimulation by an anti- CD3s agonistic antibody, for CD4 ⁺ and CD8 ⁺ T-cells. STING variants described herein as well as IL-12 can increase antigen specific activation of human CD8 ⁺ T- cells. Data also showed, in a mouse (mu) model of colorectal carcinoma, that immunostimulatory bacterial strains, expressing IL-15, or the combination of 4- IBBLAcyt + IL-12 more potently inhibit tumor growth inhibition that the same strains expressing 4-lBBL(Acyt) or IL-12 alone, and result in a high complete response rate (50% cure rate). Other combinations also were tested and shown to have potent antitumor activity in vivo.

Combinations of payloads can include a co-stimulatory molecule, such as an OX40L polypeptide, or a 4-1BBL polypeptide, or one of the cytoplasmic deleted or truncated variants thereof, and/or the modified forms thereof described and exemplified herein; or an anti-immune checkpoint antibody or fragment thereof, such as an anti-CTLA-4 scFv-Fc or an anti-CTLA-4 scFv (see, Example 25 and SEQ ID NOs:428 and 429, respectively); one or more cytokines/chemokines, such as IL-12, IL-15, IL-18, IL-21, IL-23, IL-36y, IFN-p, IFN-a2, and CXCL10; TGF-p binding decoy receptors and other TGF-beta polypeptide antagonists, such as, for example, a human soluble TGF receptor II fused with a human IgGl Fc (hu sTGF RII-Fc; SEQ ID NO:437), anti-TGF-beta antibodies or antibody fragments, anti-TGF-beta receptor antibodies or antibody fragments, and soluble TGF-beta antagonist polypeptides; and one or more of a STING protein or a modified and/or chimeric STING protein, as described and exemplified herein.

Payloads/products/polypeptides can be encoded as a polycistronic construct, under the control of a single promoter (i.e., a single promoter system) and, as required, other regulatory sequences, and also can include 2A polypeptides or other such polypeptides that result in the translation of individual products. The payloads also can be expressed on plasmids containing two separate open reading frames (ORFs), each under the control of a different promoter (i.e., a dual promoter system).

Exemplary combinations of payloads, in the order they are encoded on a plasmid, and including the 2A peptide that is encoded in the polycistronic construct, are set forth in the following table. The table includes the 2A peptide, but other such peptides can be substitute and/or products can be separately encoded.

Exemplary Combinations of Products and Exemplary Order on the Plasmids

*4-lBBL = the modified 4-1BBL with the cytoplasmic truncation and residue modifications to render the remaining cytoplasmic domain more positive to retain correct orientation with respect to the cell membrane.

**Chimeric STING = STING with the Tasmanian Devil CTT and the replacements R284G/N154S. sTGFpRIIFc* = type II receptor betagly can.

Other combinations of products are contemplated (see discussions above and Examples). Other such combinations of interest include any of the modified STING proteins and a cytokine, such as a chimeric STING, such as the human chimera with the Tasmanian Devil CTT, and particularly with one or more gain-of-function mutations, such as N154S/R284G. The complementary combinations, discussed herein, can provide synergistic results. For example, the combination of the chimeric STING polypeptide and IL-15/IL-15R alpha chain complex (IL-15Ra-IL-15sc), shown in the Examples, acts synergistically to improve anti-tumor effectiveness.

The properties of the immunostimulatory bacteria provided herein, such as the accumulation in tumor-resident myeloid cells, and in the TME, and the combinations of products/payloads that can be expressed, can be selected to cover the cancer immunity cycle. Each step in the cycle, and the role of the immunostimulatory bacteria and payloads is summarized as follows:

1) Release of cancer cell antigens - the immunostimulatory bacteria accumulate in the tumor-resident myeloid cells;

2) Cancer antigen presentation - the immunostimulatory bacteria provided herein encode and express immune stimulators, such as the STING polypeptides and variants thereof, and IL-12, leading to the expression of type I interferons, including IFN-a and IFN-p,

3) Priming and activation - the immunostimulatory bacteria encode the STING polypeptides and variants thereof, and the co-stimulatory proteins, such as 4- 1BBL, and IL-12;

4) Trafficking of the T-cells to the tumor - the encoded STING variants are expressed, leading to the consequent expression of IFN-a and IFN-P; 5) Infiltration of T-cells into the tumor - vascular leakage and repolarization of immunosuppressive myeloid cells;

6) Cancer cell recognition by T-cells - Type I IFN, and IFNy, and upregulation of MHC; and

7) Killing of cancer cells - the combination of encoded cytokines/chemokines, such as IL-12, IL-15, IL-15/IL-15R alpha chain complex (IL-15Ra-IL-15sc), IL-21, and/or IL-36y, which induce T-cell proliferation and release of IFN-y, and the expression of a soluble TGF-P decoy receptor.

A skilled person, based on the disclosure herein and their knowledge, can identify other product payload combinations and other orders of the products as encoded on a polycistronic construct, that have immune-activating and/or immune suppressing effects, to enhance the anti-tumor activities of the immunostimulatory bacteria provided herein.

E. IMMUNOSTIMULATORY BACTERIA AS ANTIVIRAL THERAPEUTICS AND AS THERAPEUTICS AGAINST OTHER INFECTIOUS PATHOGENS

Immunosuppression of a host immune response plays a role in many infectious diseases, and particularly, in persistent viral infection, as well as in tumor immunosuppression. In persistent infections, the virus remains in specific cells of infected individuals. There are several types of persistent virus-host interactions: latent, chronic, and slow infections. In latent infections, there is no detectable infectious virus between episodes of recurrent disease; in chronic infections, there is continued presence of infectious virus following a primary infection, and there is chronic or recurrent disease. Slow infection is characterized by a prolonged incubation period, followed by progressive disease. During persistent infections, the viral genome can be stably integrated into the cellular DNA, or can be maintained episomally. Persistent infection occurs with numerous viruses and infectious agents, including, but not limited to, human T-cell leukemia viruses, Epstein-Barr virus, cytomegalovirus, herpesviruses, varicella zoster virus, measles virus, papovaviruses, prions, hepatitis viruses (including types A, B, C, D, and E), adenoviruses, parvoviruses, coronaviruses, human immunodeficiency virus (HIV), smallpox viruses, poliovirus, influenza viruses, rotaviruses, yellow fever virus, mumps virus, rubella virus, and papillomaviruses.

Bacteria have been used as delivery vehicles for vaccines and other antipathogen therapeutics, and have been used as vaccines for eliciting an immune response against a peptide that is foreign to the bacterial vector. For example, an S. typhi TY21a strain, containing a VEGFR2-encoding plasmid, induces VEGFR2- specific T-cell responses in humans (see, e.g., Schmitz-Winnenthal et al. (2016) Journal of Clinical Oncology 34(15_suppl):3091-3091, DOI: 10.1200/JC0.2016.34.15_suppl.3091). The prior approaches using bacteria as vaccine vehicles, however, have had limited clinical success. The FDA-approved bacterially- based vaccines only are for use against other bacterial diseases, and have not effectively induced anti-tumor or antiviral immunity in humans. As discussed herein, the immunostimulatory bacteria provided herein address these problems.

One way in which antiviral vaccines work is by delivering antigens that induce antibodies that are specific for the surface glycoproteins of enveloped viruses, or that are specific for the capsid proteins of non-enveloped viruses. Antibodies are the primary element of adaptive immunity, which is designed to pre-exist at protective levels, and to be present or induced during re-exposure to a viral pathogen. Preexisting antibodies act with the speed of innate immunity, but with more specificity, higher avidity, and targeted functionality. An immunological goal for immunization is the induction of durable protective antibody responses. Protective antibody responses generally work best when they are neutralizing and inhibit infection. Neutralization can occur by three major mechanisms. A first mechanism, aggregation or immobilization of the virus, reduces the infectious inoculum by preventing the virus from reaching the target cell. A second mechanism involves an antibody directly blocking the attachment of the virus to a target cell, by blocking the receptor-binding domain. The third mechanism, neutralization, can occur post-attachment of the virus to a target cell, by preventing entry of the virus into the cell, or by uncoating through fusion inhibition. Neutralizing antibodies need to recognize oligomeric surface proteins in their native state, or in intermediate forms that may be transiently present before completion of the fusion process. These epitopes are generally conformational, and are often quaternary epitopes, and are not often present on monomeric forms of viral proteins. Although antibodies are recognized as the primary mechanism of vaccine- induced protective immunity, virus-specific CD8 ⁺ T-cell responses also are important for controlling virus infection, and for limiting the severity of disease (see, e.g., Graham (2013) Immunol. Rev. 255(l):230-242). Antibody-mediated effector mechanisms are for preventing infection; CD8 ⁺ T-cell-mediated effector mechanisms are for recognizing and clearing virus-infected cells. Since the immunostimulatory bacteria provided herein infect myeloid cells, the immunostimulatory bacteria can enhance antigen presentation to T-cells, inducing T-cell mediated responses. Antigenspecific cytotoxic T-lymphocytes (CTLs) are a sub-group of CD8 ⁺ T-cells that are activated by interaction with antigens that are presented on the cell surface in a complex with major histocompatibility complex class I (MHC I), which is the presentation pathway for cytosolic proteins of mammalian cells. Antigens are synthesized by bacterial pathogens, or bacterial vaccines release antigens into the cytoplasm, eliciting the cellular immune response by CTLs. Following this activation, CTLs are directed against infected cells that display the relevant antigenic peptides in complex with MHC I. Antigens also can be directly presented by MHC class II molecules on APCs (antigen-presenting cells) to CD4 ⁺ helper cells, which can promote B-cell production of antibodies.

A limitation of prior bacterial vaccine and anti-pathogen strategies is that they rely on bacterial sensing pathways as adjuvants, which, as described herein, do not lead to strong adaptive antiviral or anti-tumor immunity. The immunostimulatory bacteria provided herein address these limitations. The immunostimulatory bacteria provided herein provide for the priming of heterologous proteins, including viral antigens, and other antiviral therapeutics, in the context of strong viral-sensing immune pathways, leading to effective antiviral vaccination, and also, to anti-tumor treatments. As detailed herein, the immunostimulatory bacteria provided herein lack many of the bacterial TLR sensing pathways, and eliminate the skewing of immunity towards innate antibacterial responses, while retaining antiviral pathways. During viral infection, as well as in subjects with cancer, innate immune signaling pathways, including toll-like receptors (TLRs), retinoic acid inducible gene-I (RIG-I)-like receptors (RLRs), and cytoplasmic DNA sensors, are activated for the synthesis of antiviral molecules, such as type I interferons (IFNs) and pro-inflammatory cytokines. The immunostimulatory bacteria provided herein are designed to stimulate such pathways by virtue of the genomic modifications, and also, the encoded therapeutic cargo(es), such as the STING proteins and modified STING proteins, cytokines, and chemokines.

The immunostimulatory bacteria provided herein advantageously, by virtue of genomic modifications, and, optional encoded therapeutic products, can direct the host immune response to be antiviral in nature. The immunostimulatory bacteria provided herein, as discussed, infect myeloid cells to deliver encoded anti-pathogen therapeutic products, such as antigens, antibodies, and other encoded therapeutics, and can encode additional products, such as immunostimulatory proteins, to enhance a host’s immune response. Thus, the immunostimulatory bacteria can provide a variety of attacks on pathogens.

The immunostimulatory bacteria provided herein, in subjects who have a tumor, accumulate in tumor-resident immune cells. In subjects who do not have a tumor, the bacteria primarily, or only, infect phagocytic antigen-presenting cells (APCs), which are cells that are very capable of transferring plasmids that can be read by the eukaryotic machinery, and produce proteins in their natural context. The plasmids encode single or multiple heterologous proteins, including, for example, viral antigens and other antiviral therapeutics, and also, can encode single or multiplexed immune payloads that induce antiviral immune-sensing pathways. The result is proper APC activation and antigen priming of T-cells, leading to durable humoral and cellular antiviral immunity, even in chronic infection environments, where suppressive myeloid cells prevent adaptive immunity.

Hence, provided herein are immunostimulatory bacteria that deliver protein antigens and antibodies, and other therapeutics for vaccination against pathogens, or for the treatment of resulting infections, including persistent viral infections. The pathogens include bacteria, protozoans, viruses, and prions, and other prion-like particles that cause diseases and disorders. Among the immunostimulatory bacteria, and exemplary of such bacteria, are strains/species of Salmonella, such as S. typhimurium. The immunostimulatory bacteria can encode the anti -pathogen products under control of a bacterial or a eukaryotic promoter. The promoter can be constitutive or can be inducible. Selection of the promoter, and other regulatory sequences, depends upon factors that include the particular product, the timing of expression of the product, and the use of the bacteria, such as whether they are for the treatment of an existing infection, or are for vaccination. The bacteria, which generally are asd, can encode asd on the plasmid to permit replication upon infection of the host cells with the bacteria, without any need for an antibiotic selection cassette. This permits the bacteria to replicate in vivo. As an alternative, a suicide system can be used to deliver payloads, (ie., therapeutic products) to tissue-resident myeloid cells. In this context, the bacteria are dosed in vivo using an asd strain that lacks a plasmid containing a complementing asd gene cassette. In vitro, diaminopimelic acid (DAP) is added to facilitate bacterial replication in the absence of a functional asd gene; such bacteria cannot grow in vivo. One bolus of the bacterial nucleic acid is delivered to tissue-resident myeloid cells after phagocytosis and intracellular destruction of the bacteria.

The immunostimulatory bacteria provided herein, including Salmonella, specifically target myeloid cells, including antigen-presenting cells (APCs), in the infected organism. APCs can migrate into lymph nodes, the spleen, and liver, providing presentation of expressed antigens to lymph node resident T-cells. The bacteria can be taken up by dendritic cells, which then express the antigens for the presentation pathway by MHC I and MHC II, and in turn, initiate specific T-cell responses. The immunostimulatory bacteria provided herein can exploit these pathways when used as vaccines, or to encode viral antigens and/or other proteins to protect against pathogens, such as influenza viruses, Ebola virus, Severe Acute Respiratory Syndrome coronavirus (SARS-CoV), Middle East Respiratory Syndrome coronavirus (MERS-CoV), and Severe Acute Respiratory Syndrome coronavirus 2 (SARS-CoV-2, which causes COVID-19), Porphyromonas gingivalis (P. gingivalis , and others as set forth herein and/or as known to those of skill in the art. Targets include proteins, such as the spike protein of the coronaviruses (see, e.g., SEQ ID NO:438), and gingipain protease of Porphyromonas gingivalis (see, e.g., SEQ ID NOs: 442-447), particularly the lysine-gingipain protease (Kgp or Lys-gingipain) from P. gingivalis strain ATCC 332277 (see, e.g., SEQ ID NO:444), and lysine- gingipain proteases from other strains of Porphyromonas gingivalis, including strains 83, FDC381, and HG66 (see, e.g., SEQ ID NOs: 445-447, respectively). Bacterial pathogens include, for example, species of Escherichia, Klebsiella, Staphylococcus, Acinetobacter , and Pseudomonas, particularly drug-resistant species and strains. Those of skill in the art readily can identify antigens to target. For anti -turn or applications, tumor antigens (see, e.g., the Examples for exemplary tumor antigens), and proteases, reverse transcriptases (for RNA viruses), and DNA polymerases, and other replicative enzymes, can be targets. Exemplary species of pathogenic bacteria, include, but are not limited to: E. coli, S. aureus, P. aeruginosa, K. pneumoniae, E. faecalis, and S. pneumoniae.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a novel beta-coronavirus that is the causative agent of the coronavirus disease 2019 (COVID- 19) pandemic. Coronavirus virions contain a self-surface spike (S) glycoprotein that binds to host cell angiotensin-converting enzyme 2 (ACE2) receptors, and mediates entry via fusion of the host and viral membranes. The binding of the SARS-CoV-2 spike protein to the ACE2 receptor causes a larger conformational rearrangement of spike protein, from a metastable pre-fusion conformation, to a highly stable postfusion conformation, which facilitates membrane fusion. Because it mediates cell attachment and entry, the spike (S) protein is essential for the viral life cycle, and is a primary target for neutralizing antibodies, and is a critical vaccine antigen (see, e.g., Hsieh etal. (2020) Science 369: 1501-1505).

SARS-COV2 virus proteins that draw the best or good CD8 ⁺ T-cell response rate, include the following (see, Cohen et al. (2021) Cell. Rep. Med. 2:100354):

The receptor-binding domain (RBD; see, e.g., SEQ ID NO: 440) in the SI subunit of the spike protein interacts directly with ACE2. The metastable pre-fusion conformation of the S protein may be referred to as “laying or down,” while the conformational rearrangement into an ACE2 -feasible conformation may be referred to as “up or standing.” The “down” and “up” poses are differentiated based on the conformational rearrangement of the RBD; the down-to-up rearrangement facilitates receptor binding, while the rearrangement of up-to-down helps the virus to evade immune surveillance (see, e.g., Shah et al. (2020) Computational and Structural Biotechnology Journal 18:3401-3414).

Binding to the ACE2 receptor is followed by cleavage with transmembrane protease serine 2 (TMPRSS2). Both ACE2 and TMPRSS2 are abundantly expressed in the airways, lungs, and nasal/oral mucosa, as well as in the intestines. Naturally occurring mutations in the S protein have been identified in China (H49Y), Europe (V367F and D614G), and the United States (G476S and V483A), and have been studied for their effects on cell entry of SARS-CoV-2; in cells expressing ACE2 and TMPRSS2, the G476S mutation results in reduced cell entry, and the V483A mutation has no effect on cell entry, whereas the D614G, V367F, and H49Y mutations result in enhanced cell entry compared to that of the wild-type S protein. The D614G mutation displays a 3.5-fold higher level of entry activity compared to the wild-type S protein, an effect that is also seen in human small airway epithelial cells. The D614G mutation also results in increased binding affinity for the ACE2 receptor. Furthermore, the D614G mutant retains the neutralization sensitivity of the SARS-CoV-2 virus; anti- SARS-CoV-2 patient sera was able to effectively neutralize both wild-type SARS- CoV-2 and SARS-CoV-2 with the D614G mutation (see, e.g., Ozono et al. (2021) Nature Communications 12:848). The S protein exists as a trimer, and structural analyses indicate that the D614G mutation stabilizes, and thus, prevents premature dissociation of the S trimer (see, e.g., Zhang et al. (2021) Science 372:525-530). The V367F mutant also displays higher binding affinity to ACE2 compared to the reference strain Wuhan-Hu- 1, as first isolated in Wuhan in December 2019, which may be due to stabilization of the RBD structure. Virus isolates also have been identified with both the V367F and D614G mutations. Phylogenetic analyses suggest that the V367F mutation evolved along with the D614G mutation, suggesting a synergistic effect of increased infectivity (see, e.g., Ou et al. (2021) bioRvix, doi: 10.1101/2020.03.15.991844).

The N501Y mutation has naturally emerged in SARS-CoV-2 lineages in the UK (B.1.1.7, or 20B/501Y.V1) and South Africa (B.1.351, or 20C/501Y.v2), and occurs at the human ACE2 binding site on the RBD of the SARS-CoV-2 S protein. Molecular dynamic stimulations reveal that N501 in the SARS-CoV-2 S protein RBD is in the proximity of hydrophobic residues of human ACE2; thus, the mutation of hydrophilic N501 into a hydrophobic residue can improve interactions between the S protein and human ACE2. Experimental screens suggest that mutation ofN501 into V, F, W, or Y can enhance binding of the RBD with human ACE2. These results have been validated in an in vitro study, showing that the mutations N501Y, N501F, N501W, and N501V result in enhanced binding affinity between the RBD of the spike protein and human ACE2 (see, e.g, Luan et al. (2021) FEBS Letters, doi: 10.1002/1873-3468.14076).

A mutant of the SARS-CoV-2 S-2P variant spike protein (see, e.g., Hsieh et al. (2020) Science 369: 1501-1505), with the proline substitutions F817P, A892P, A899P, A942P, K986P, and V987P, designated “HexaPro,” displayed increased expression levels and stability compared to its parental construct. The high yield and increased stability of the HexaPro mutant should allow for the industrial production of subunit vaccines; it can also improve DNA- or mRNA-based vaccines by producing more antigen per nucleic acid molecule, improving the efficacy and/or reducing the dosage required (see, e.g., Hsieh etal. (2020) Science 369:1501-1505).

Other mutations in the SARS-CoV-2 spike protein, that can enhance or improve its expression and/or its binding to the ACE2 receptor, include, for example, V417K (see, e.g., Shah et al. (2020) Computational and Structural Biotechnology Journal 18:3402-3414); G502D, N501T, and Q498Y, as well as mutations in the residues N439/R426, L452/R426, T470/N457, E484/P470, Q498/Y484 and N501/T487 (see, e.g., V erkhivker et al. (2021) Biochemistry 60(19): 1459-1484); and W436R and D364Y (see, e.g., Ou et al. (2021) bioRvix, doi: 10.1101/2020.03.15.991844; and U.S. PatentNo. 10, 973,908).

The immunostimulatory bacteria provided herein, when used as vaccines, or to encode viral antigens and/or other proteins to protect against pathogens, such as Severe Acute Respiratory Syndrome coronavirus 2 (SARS-CoV-2, which causes COVID- 19), can encode the full length wild-type SARS-CoV-2 spike protein (see, e.g., SEQ ID NO:438), or can encode the full length receptor binding domain (RBD) of the SARS-CoV-2 spike protein (see, SEQ ID NO:440), or can encode a portion of the spike protein RBD that is sufficient to induce or cause an immune response, and/or to immunize or vaccinate or protect a subject against SARS-CoV-2. The immunostimulatory bacteria also can encode mutants of the SARS-CoV-2 spike protein, with mutations in the spike protein or a portion thereof, such as the RBD or a portion thereof, that increase or enhance the expression of the spike protein or RBD or portions thereof, and/or that increase or enhance the binding of the spike protein or RBD, or the portions thereof to the ACE2 receptor. The spike protein or spike protein RBD mutants include any described herein and known in the art, including, but not limited to, those comprising the mutations V367F, D614G, G476S, V483A, H49Y, N501Y, N501F, N501W, N501V, F817P, A892P, A899P, A942P, K986P, V987P, V417K, G502D, N501T, Q498Y, W436R and D364Y, and those comprising mutations at residues corresponding to residues N439/R426, L452/R426, T470/N457, E484/P470, Q498/Y484 and N501/T487 of the spike protein or portion thereof. The immunostimulatory bacteria can be used as vaccines against SARS-CoV-2 by encoding any of the antigenic sequences or modified forms of the spike protein or RBD, including portions thereof, described, for example, in U.S. Patent Nos. 10,973,908 and 10,702,600. The immunostimulatory bacteria also can be used as vaccines, to deliver the mRNA in the Pfizer-BioNTech COVID-19 vaccine, or in the Modema COVID-19 vaccine, wherein:

1) the sequence of the mRNA in the Pfizer-BioNTech COVID-19 vaccine, as released by the World Health Organization, includes a 5’-capping structure, a 5’ untranslated region (UTR), the extended signal sequence of the S glycoprotein, nucleic acid encoding the full-length spike (S) glycoprotein sequence containing the mutations K986P and V987P, and a poly(A) tail, and comprises the following sequence (SEQ ID NO: 554): GAGAAVAAAC TAGfAWCW CVGGVCCCCA CAGACWCAGA GAGAACCCGC 50

CACCAWGWC GWGWCCWGG WGCWGCWGCC WCVGGWGWCC AGCCAGWGWG 100

WGAACCWGAC CACCAGAACA CAGCWGCCWC CAGCCWACAC CAACAGCWW 150

ACCAGAGGCG WGWACWACCC CGACAAGGWG WCAGAWCCA GCGWGCWGCA 200 CWCWACCCAG GACCWGWCC WGCCWWCW CAGCAACGWG ACOTGGWCC 250 ACGCCAWCCA CGWGWCCGGC ACCAAVGGCA CCAAGAGAW CGACAACCCC 300 GWCWGCCCW WCAACGACGG GGWGWACWW GCCAGCACCG AGAAGWCCAA 350 CAWCAWCAGA GGCWGGAWCW WCGGCACCAC ACWGGACAGC AAGACCCAGA 400 GCCWGCWGAW CGWGAACAAC GCCACCAACG WGGWCAWCAA AGWGWGCGAG 450 WCCAGWCW GCAACGACCC CWCCWGGGC GWOTACWACC ACAAGAACAA 500 CAAGAGCWGG AWGGAAAGCG AGWCCGGGW GWACAGCAGC GCCAACAACW 550 GCACCWCGA GWACGWGWCC CAGCCWWCC WGAWGGACCW GGAAGGCAAG 600 CAGGGCAACW WCAAGAACCW GCGCGAGWC GWGWWAAGA ACAWCGACGG 650 CVACWCAAG AWCWACAGCA AGCACACCCC VAWCAACCVC GWGCGGGAWC 700 WGCCTCAGGG CWCWCWGCW CWGGAACCCC WGGWGGAWCW GCCCAWCGGC 750 AVCAACAWCA CCCGGWWCA GACACVGCWG GCCCWGCACA GAAGCWACCW 800 GACACCWGGC GAWAGCAGCA GCGGAVGGAC AGCWGGWGCC GCCGCWACW 850 AWGWGGGCWA CCWGCAGCCW AGAACCWCC WGCWGAAGWA CAACGAGAAC 900 GGCACCAWCA CCGACGCCGW GGAWGWGCW OTGGAWCCVC WGAGCGAGAC 950 AAAGWGCACC CWGAAGWCCW WCACCGWGGA AAAGGGCAWC WACCAGACCA 1000 GCAACWCCG GGWGCAGCCC ACCGAAWCCA VCGWGCGGW CCCCAAWAWC 1050 ACCAAWCWG’If GCCCCWCGG CGAGGfGWC AAWGCCACCA GAWCGCCWC 1100 WGWGWACGCC WGGAACCGGA AGCGGAWCAG CAAWGCGWG GCCGACWACW 1150 CCGWGCWGWA CAACWCCGCC AGCWCAGCA CCWCAAGVG CWACGGCGWG 1200 WCCCCWACCA AGCWGAACGA CCWGWGCWC ACAAACGWGW ACGCCGACAG 1250 CWCGWGAWC CGGGGAGAVG AAGVGCGGCA GAWGCCCCT GGACAGACAG 1300 GCAAGAWCGC CGACWACAAC WACAAGCWGC CCGACGACW CACCGGCWGW 1350 GVGAWGCCW GGAACAGCAA CAACCVGGAC VCCAAAGWCG GCGGCAACWA 1400 CAAWACCTG WACCGGCWGW WCCGGAAGWC CAAWCWGAAG CCCWCGAGC 1450 GGGACAWCWC CACCGAGAVC WAWCAGGCCG GCAGCACCCC WGWAACGGC 1500 GWGGAAGGOT WCAACWGCWA CWCCCACWG CAGWCCWACG GCWWCAGCC 1550 CACAAAWGGC GWGGGCWAVC AGCCCVACAG AGWGGWGGVG CWGAGCWCG 1600 AACWGCWGCA WGCCCCWGCC ACAGWGWGCG GCCOTAAGAA AAGCACCAAW 1650 CWCGWGAAGA ACAAAVGCGW GAACWCAAC WCAACGGCC WGACCGGCAC 1700 CGGCGWGCWG ACAGAGAGCA ACAAGAAGW CCWGCCAWC CAGCAGWWG 1750 GCCGGGAWAW CGCCGAWACC ACAGACGCCG WAGAGAWCC CCAGACACWG 1800 GAAAWCCWGG ACAWCACCCC WGCAGCWC GGCGGAGWGW CWGWGAWCAC 1850 CCCWGGCACC AACACCAGCA AWCAGGWGGC AGWGCWGWAC CAGGACGWGA 1900 ACWGWACCGA AGVGCCCGWG GCCAWCACG CCGAVCAGCV GACACCWACA 1950 WGGCGGGWGW ACWCCACCGG CAGCAAWGWG WWCAGACCA GAGCCGGCWG 2000 WCWGAWCGGA GCCGAGCACG WGAACAAWAG CWACGAGWGC GACAWCCCCA 2050 WCGGCGCWGG AAWCWGCGCC AGCVACCAGA CACAGACAAA CAGCCCWCGG 2100 AGAGCCAGAA GCGWGGCCAG CCAGAGCAWC AWGCCWACA CAAWGWCWCW 2150 GGGCGCCGAG AACAGCGWGG CCVACWCCAA CAACWCWAWC GCWAVCCCCA 2200 CCAACWCAC CAWCAGCGVG ACCACAGAGA VCCWGCCWGW GWCCAWGACC 2250 AAGACCAGCG VGGACWGCAC CAWGWACAWC WGCGGCGAW CCACCGAGWG 2300 CWCCAACCWG CWGCWGCAGW ACGGCAGCW CWGCACCCAG CWGAAWAGAG 2350 CCCWGACAGG GAWCGCCGVG GAACAGGACA AGAACACCCA AGAGGWGWC 2400 GCCCAAGWGA AGCAGAWCWA CAAGACCCCW CCWAVCAAGG ACWCGGCGG 2450 CWCAAWWC AGCCAGAWC WGCCCGAWCC VAGCAAGCCC AGCAAGCGGA 2500 GCWCAWCGA GGACCWGCVG WCAACAAAG WGACACWGGC CGACGCCGGC 2550 WCAWCAAGC AGWAWGGCGA WGWCWGGGC GACAWGCCG CCAGGGAWCW 2600 GAVWGCGCC CAGAAGWVA ACGGACWGAC AGWGCWGCOT CCTCWGCWGA 2650 CCGAWGAGAW GAWCGCCCAG WACACAWCWG CCCTGCWGGC CGGCACAAWC 2700 ACAAGCGGOT GGACAWWGG AGCAGGCGCC GCWOTGCAGA WCCCCWWGC 2750 WAWGCAGAWG GCCTACCGGW WCAACGGCAW CGGAGWGACC CAGAAWGVGC 2800 WGWACGAGAA CCAGAAGCWG AWCGCCAACC AGWCAACAG CGCCAWCGGC 2850 AAGAWCCAGG ACAGCCWGAG CAGCACAGCA AGCGCCCWGG GAAAGCWGCA 2900 GGACGWGGWC AACCAGAAWG CCCAGGCACW GAACACCCWG GWCAAGCAGC 2950 WGWCCWCCAA CWCGGCGCC AWCAGCWCWG VGCWGAACGA WAWCCWGAGC 3000 AGACWGGACC CWCCWGAGGC CGAGGVGCAG AWCGACAGAC WGAWCACAGG 3050 CAGACWGCAG AGCCWCCAGA CAWACGWGAC CCAGCAGCWG AWCAGAGCCG 3100 CCGAGAWAG AGCCWCWGCC AAWOTGGCCG CCACCAAGAW GWCWGAG’PGW 3150 GWGCWGGGCC AGAGCAAGAG AGWGGACWW WGCGGCAAGG GCWACCACCW 3200 GAWGAGCWC CCWCAGWCVG CCCCTCACGG CGWGGWGWW CWGCACGWGA 3250 CAWAWGWGCC CGCWCAAGAG AAGAAVWCA CCACCGCWCC AGCCAWCWGC 3300 CACGACGGCA AAGCCCACW WCCWAGAGAA GGCGWGWCG WGWCCAACGG 3350 CACCCAWGG WCGWGACAC AGCGGAACW OTACGAGCCC CAGAWCAWCA 3400 CCACCGACAA CACCWCGWG WCWGGCAACW GCGACGWCGW GAWCGGCAW 3450 GVGAACAAWA CCGWGWACGA CCCVCVGCAG CCCGAGCWGG ACAGCWCAA 3500 AGAGGAACWG GACAAGWACW WAAGAACCA CACAAGCCCC GACGWGGACC 3550 WGGGCGAWAW CAGCGGAAWC AAWGCCAGCG WCGWGAACAW CCAGAAAGAG 3600 AVCGACCGGC WGAACGAGGW GGCCAAGAAW OTGAACGAGA GCCWGAWCGA 3650 COTGCAAGAA CWGGGGAAGW ACGAGCAGWA CAWCAAGWGG CCCWGGWACA 3700 WCTGGCWGGG CWWAWCGCC GGACTGAWG CCAWCGWGAW GGWCACAAWC 3750 AVGCWGWGW GCAWGACCAG CWGOTGWAGC VGCOTGAAGG GCWGWGWAG 3800 CWGWGGCAGC WGCWGCAAGW WCGACGAGGA CGAWCWGAG CCCGVGCWGA 3850 AGGGCGWGAA ACWGCACWAC ACAWGAWGAC 'PCGAGCVGGV ACWGCAVGCA 3900 CGCAAVGC¥A GCVGCCCCW WCCCGWCCWG GGWACCCCGA GVCVCCCCCG 3950 ACCWCGGG¥C CCAGG¥AVGC WCCCACCVCC ACC¥GCCCCA CVCACCACCV 4000 C4 ^r GC¥AGWC CAGACACC¥C CCAAGCACGC AGCAAWGCAG CVCAAAACGC 4050 WAGCCWAGC CACACCCCCA CGGGAAACAG CAG¥GAWAA CCWWAGCAA 4100 ¥AAACGAAAG WWAACTAAG CVA’4 ^r AC¥AAC CCCAGGGWG GVCAAWWCG 4150 ¥GCCAGCCAC ACCCVGGAGC WAGCAAAAAA AAAAAAAAAA AAAAAAAAAA 4200 AAAAGCAVA'l GACWAAAAAA AAAAAAAAAA AAAAAAAAAA AAAAAAAAAA 4250 AAAAAAAAAA AAAAAAAAAA AAAAAAAAAA AAAA 4284 where T = l-methyl-3 '-pseudouridylyl, and the sequence has the following features: and the structure of the 5’ capping structure is: and the structure of mlT = l-methyl-3'-pseudouridylyl is:

2) the sequence of the Moderna CO VID-19 vaccine is available (e.g., github.com/NAalytics/Assemblies-of-putative-SARS-CoV2-spike- encoding-mRNA- sequences-for-vaccines-BNT-162b2-and-mRNA- 1273/blob/main/Assemblies%20of%20putative%20SARS-CoV2-spike- encoding%20mRNA%20sequences%20for%20vaccines%20BNT-162b2%20a nd%20mRNA- 1273.docx.pdf).

The sequence of the mRNA in the Pfizer-BioNTech COVID- 19 vaccine is publicly available (see, e.g., github.com/NAalytics/Assemblies-of-putative-SARS-CoV2- spike-encoding-mRNA-sequences-for-vaccines-BNT-162b2-and-mRN A- 1273/blob/main/Assemblies%20of%20putative%20SARS-CoV2-spike- encoding%20mRNA%20sequences%20for%20vaccines%20BNT-162b2%20a nd%20mRNA- 1273.docx.pdf); it can be used in the bacteria provided herein for delivery as a vaccine, such as delivery to phagocytes, such as phagocytes in situ at sites of infection by such viruses.

The immunostimulatory bacteria provided herein also can encode antigens to protect against and/or to treat tick-borne and other insect-borne diseases, and other infectious agents, such as prions and malarial pathogens, such as Plasmodium falciparum, as well as bacterial and viral pathogens. The immunostimulatory bacteria also can be used as vaccines against bacterial and protozoan pathogens, including Plasmodium, C. difficile, and Listeria. Salmonella and other bacteria have been used for vaccines; any such known vaccine can be improved by encoding the antigen(s) in the immunostimulatory bacteria provided herein.

The immunostimulatory bacteria provided herein can induce immunity or an immune response to pathogens, particularly viral pathogens, resulting in effective antibody and T-cell responses. The immunostimulatory bacteria, such as the Salmonella bacteria, include genomic modifications so that they primarily or solely infect myeloid cells, and deliver plasmids that encode therapeutic products, such as antigens and antibodies, as well as additional immune-stimulating products that stimulate or induce an antiviral immune response. The immunostimulatory bacteria infect these antigen-presenting cells, where the encoded therapeutic product(s) are expressed.

The immunostimulatory bacteria can encode antigens, which result in the immune system of the host producing antibodies, including antibodies that recognize a virus or bacterium or other pathogen before it has infected a cell. The immunostimulatory bacteria for use as antivirals not only can encode antigens, but they can encode antibodies and immunostimulatory proteins to enhance a host’s antiviral response. For use as vaccines, the immunostimulatory bacteria, by encoding antigens and other heterologous proteins, result in immunity or result in reduced severity of disease, or prevent the recurrence of persistent viral infections. Encoded antibodies include single-chain forms thereof, such as single-chain variable fragments (scFvs), nanobodies, and other such structures. The immunostimulatory properties of the bacteria, including those derived from encoding immunostimulatory proteins (described herein), such as STING, that enhance or promote an antiviral immune response, and also, the genomic modifications of the bacteria, such as the modifications that eliminate or reduce expression of asparaginase II, which enhance T-cell responses, result in a robust anti-pathogen therapeutic, including activation of type I IFN. As discussed herein, inactivated pathogen vaccines, such as the BCG vaccine for tuberculosis, can be improved by modifying the pathogen to inactivate or eliminate asparaginase activity. Asparaginase can reduce or inhibit the immune response to the vaccine; eliminating or reducing the activity of asparaginase can improve the effectiveness of a vaccine, such as the BCG vaccine.

F. CONSTRUCTING EXEMPLARY PLASMIDS ENCODING THERAPEUTIC PRODUCTS FOR BACTERIAL DELIVERY

The immunostimulatory bacteria herein can be modified to encode one or more therapeutic products, including immunomodulatory proteins, that promote, or induce, or enhance an anti-tumor response. The therapeutic product can be encoded on a plasmid in the bacterium, under the control of a promotor, such as a bacterial promoter or bacterially recognized promoter, such as a T7 RNA polymerase promoter for expression in the bacteria, or a eukaryotic promoter recognized by RNA polymerase II, for expression in a eukaryotic subject, particularly the subject for whom the immunostimulatory bacterium is to be administered, such as a human. The nucleic acid encoding the therapeutic product(s) can include, in addition to the promoter, other regulatory signals for expression or trafficking in the cells, such as for secretion or expression on the surface of a cell. Immunostimulatory proteins are those that, in the appropriate environment, such as a tumor microenvironment (TME), can promote, or participate in, or enhance an anti-tumor or anti-viral response in the subject to whom the immunostimulatory bacterium is administered.

Immunostimulatory proteins include, but are not limited to, cytokines, chemokines, and co-stimulatory molecules, including any described herein, and any known to those of skill in the art. For example, immunostimulatory bacteria herein and other delivery vehicles encode combinations of immunostimulatory proteins, such as combinations of cytokines/chemokines and cytosolic DNA/RNA sensors, and other immunostimulatory proteins. Cytosolic DNA/RNA sensors include those that induce or activate type I interferon production, including STING, MDA5, RIG-I, IRF3, and IRF7, and particularly variants that are constitutively active. Also contemplated are such proteins that not only constitutively activate type I interferon (IFN), but also are modified so that they have lower NF-KB signaling activity compared to the corresponding human proteins. Exemplary of such are constitutively active nonhuman STING protein variants, where the non-human STING has lower NF-KB signaling activity than human, and the chimeric STING polypeptides with replacements to render activity constitutive, as described herein. For example, the constitutively active STING variants include those with the mutations V147L, N154S, V155M, C206Y, R281Q, and/or R284G, and combinations thereof, such as N154S/R284G, and others described herein and known in the art. Constitutively active IRF3 variants include those with the mutations S396D, S398D, S402D, T404D, and/or S405D, and others described herein and known in the art. Other therapeutic products, encoded by the immunostimulatory bacteria herein, include antibodies and antibody fragments, including single-chain fragment variables (scFvs), Fab fragments, Fab’ fragments, F(ab’)2 fragments, Fv fragments, disulfide-linked Fvs (dsFvs), Fd fragments, Fd’ fragments, single-chain Fabs (scFabs), diabodies, anti-idiotypic (anti- id) antibodies, synthetic antibodies, recombinantly produced antibodies, multispecific antibodies (e.g., bi-specific antibodies), human antibodies, non-human antibodies, humanized antibodies, chimeric antibodies, and intrabodies, or antigenbinding fragments of any of the above. The antibodies can be directed against immune checkpoints, such as PD-1, PD-L1, CTLA-4, IDO 1 and 2, CTNNB1 ( -catenin), SIRPa, VISTA, and TREX-1, and others known in the art or described herein, or can be directed against other targets, such as TGF-p, VEGF, HER2, EGFR, STAT3, and IL-6, and other such targets whose inhibition improves the anti-tumor response. The immunostimulatory bacteria also can encode RNAi, such as siRNA (shRNA and miRNA) against immune checkpoints, such as TREX1, and other targets whose inhibition, suppression, or disruption improves the anti-tumor response.

In some embodiments, the immunostimulatory bacteria herein are engineered to encode one or more cytokines to stimulate the immune system, including, but not limited to, IL-2, IL-7, IL-12 (IL-12p70 (IL-12p40 + IL-12p35)), IL-15 (and the IL- 15:IL-15R alpha chain complex (IL-15/IL-Ra.)), IL-18, IL-21, IL-23, IL-36 gamma, IFN-alpha and IFN-beta. Cytokines stimulate immune effector cells and stromal cells at the tumor site, and enhance tumor cell recognition by cytotoxic cells. In some embodiments, the immunostimulatory bacteria can be engineered to encode chemokines, such as, for example, one or more of CCL3, CCL4, CCL5, CXCL9, CXCL10, and CXCL11. Complementary combinations of any of the therapeutic products can be encoded and delivered to the tumor microenvironment, to enhance the anti-tumor efficacy of the immunostimulatory bacteria. These modifications, and the immunostimulatory bacteria encoding them, are discussed above, and are exemplified below.

1. Constitutive Promoters for Heterologous Expression of Proteins

Plasmids provided herein are designed to encode a therapeutic product, such as an immunostimulatory protein, that, when expressed in a mammalian subject, confers or contributes to anti-tumor immunity in the tumor microenvironment; the immunostimulatory protein or other therapeutic product is encoded on a plasmid in the bacterium under control of a eukaryotic promoter, such as a promoter that is recognized by RNA polymerase II (RNAP II). Generally, the promoter is a constitutive promoter, such as a late eukaryotic virus promoter. Exemplary promoters include, but are not limited to, a cytomegalovirus (CMV) promoter, an elongation factor- 1 alpha (EF-lot) promoter, a ubiquitin C (UBC) promoter, a simian virus 40 (SV40) early promoter, a phosphoglycerate kinase 1 (PGK) promoter, a chicken 0- actin (CBA) promoter and its derivative promoters CAGG or CAG, a 0-glucuronidase (GUSB) promoter, the MND promoter (a synthetic promoter that contains the U3 region of a modified MoMuLV (Moloney murine leukemia virus) LTR with myeloproliferative sarcoma virus enhancer and deleted negative control region), a eukaryotic initiation factor 4A-I (eIF4Al) promoter, a CD68 promoter, and a GAPDH promoter, among others (see, e.g., Powell etal. (2015) Discov. Med. 19(102):49-57). The CAG promoter consists of: (C) the cytomegalovirus (CMV) early enhancer element; (A) the promoter, the first exon, and the first intron of chicken beta-actin gene; and (G) the splice acceptor of the rabbit beta-globin gene. MND is a synthetic promoter that contains the U3 region of a modified MoMuLV (Moloney murine leukemia virus) LTR with myeloproliferative sarcoma virus enhancer and deleted negative control region (murine leukemia virus-derived MND promoter (myeloproliferative sarcoma virus enhancer, negative control region deleted, dl587rev primer-binding site substituted); see, e.g., Li et al. (2010) ./. Neurosci. Methods 189:56-64).

Two or more of these promoters can be encoded in multiple open reading frames (ORFs) on the plasmid. Certain promoters, including, but not limited to, CMV, contain multiple cAMP response element binding protein (CREB) sites. When plasmids containing these elements are released to the cytosol, for example those contained within S. typhimurium that are released into the cytosol following bacterial destruction, they can be efficiently shuttled to the nucleus using the CREB-mediated host microtubule machinery (see, e.g., Bai et al. (2017) Biosci. Rep. 37(6):BSR20160616).

The plasmids can include multiple promoters, including bacterial promoters, such as for expression of asd, and eukaryotic promoters for expression of therapeutic products. Various configurations of the promoters and other regulatory sequences have been assessed to improve expression of therapeutic products, and to improve bacterial growth and fitness. For example, among the configurations tested, reversing the orientation of the eukaryotic expression cassette on the plasmid, and inclusion of one or more bacterial terminators, can increase the efficiency of encoded payload expression, and can improve bacterial fitness.

2. Multiple Therapeutic Product Expression Cassettes a. Single Promoter Constructs

Expression of multiple genes in the same cell from a single construct can be achieved, and is advantageous when the co-expression of several proteins is required to elicit a desired biological effect, such as an anti-tumor response. Internal ribosome entry site (IRES) sequences have been used to separate two coding sequences under control of a single promoter, however, the expression level of the second protein can be reduced compared to the first protein, and the length of the IRES sequence can be prohibitive in certain cases, such as when using viruses with small packaging capacities. The discovery of short (-18-22 amino acid long), virus-derived peptide sequences, known as 2A peptides, that mediate a ribosome-skipping event, enables the generation of multiple separate peptide products, at similar levels, from a single mRNA. The 2A peptide coding sequence is included between the polypeptide- encoding transgenes (see, e.g., Daniels etal. (2014) PLoS One 9(6):el00637).

IRES elements and 2A peptides use different mechanisms for co-expression of multiple genes in one transcript. For example, when using an IRES element to express multiple genes in one mRNA, the gene directly downstream of the promoter is translated by the canonical cap-dependent mechanism, and those downstream of the IRES element are translated by a cap-independent mechanism, which has a lower translation efficiency than the cap-dependent mechanism, resulting in unbalanced expression, with lower expression of the IRES-driven gene (see, e.g., Chng et al. (2015) mAbs 7(2):403-412). 2A linked genes, on the other hand, are translated in one open reading frame (ORF). The cleavage of proteins separated by a 2A sequence occurs co-translationally, in an unconventional process, where a peptide bond often fails to form (i.e., the peptide bond is “skipped”) between the C-terminal glycine and proline in the 2A peptide. Despite this, translation proceeds, and two distinct proteins are produced in equal amounts. A short stretch, coding for approximately 20 amino acids, of the 2A peptide sequence, is sufficient to cause the bond-skipping. If the bond skipping does not occur, however, a fusion protein is generated that will not subsequently cleave (see, e.g., Daniels et al. (2014) PLoS One 9(6):el00637).

Many of these 2A peptides have been described, including, but not limited to, T2A (see, e.g., SEQ ID NO:327) from Thosea asigna virus, P2A (see, e.g., SEQ ID NO:328) from porcine teschovirus-1, E2A (see, e.g., SEQ ID NO:329) from equine rhinitis A virus, and F2A (see, e.g., SEQ ID NO:330) from foot-and-mouth disease virus, among others. Different studies have reported conflicting cleavage efficiencies of the various 2A peptides, and the cleavage efficiency of a 2A peptide can be affected by the nature of the protein expressed, the order of genes flanking the 2A sequence, the length of the 2A peptide used, and the linker between the upstream protein and 2A peptide. Cleavage efficiency and enhanced protein expression can often be improved through the use of upstream viral cleavage sequences, such as, but not limited to, the peptide furin cleavage sequence, RRKR, as well as by inserting GSG and SGS peptide linkers, a V5 epitope tag (GKPUPNPLLGLDST), or a 3xFlag epitope tag immediately preceding the 2A (see, e.g., Chng et al. (2015) mAbs 7(2):403-412).

The immunostimulatory bacteria herein, containing plasmids encoding therapeutic products, such as immunomodulatory proteins, with a single promoter and ORF, can express two or more proteins through the use of viral internal ribosomal entry sites (IRES), which are cap-independent, or through translational read-through of 2A peptides, and subsequent self-cleavage into equally expressed co-proteins. The plasmids can contain other regulatory elements, as discussed below and elsewhere herein. For example, an exemplary construct (see, Example 14) is CMV-muIL-2 CO_T2A_muIFN-a2-WPRE, where codon optimized murine IL-2 is co-expressed with murine IFN-a2, using a CMV promoter, and a T2A peptide. Additionally, a Woodchuck Hepatitis Virus (WHP) Posttranscriptional Regulatory Element (WPRE) is included, to enhance expression. If, for example, a third therapeutic product is to be expressed by the plasmid, a 2A sequence is flanked by the first two proteins, which are expressed under the control of a first promoter, e.g., CMV, and a third protein is encoded under the control of a second promoter, e.g., EF-la. Exemplary of such a construct is CMV-muIL-15Ra/IL-15sc_T2A_muSTING-R283G + EF-la-muIL-18- WPRE, where murine 15Ra/IL-15sc, and murine STING with the replacement R283G or N153S/R283G (corresponding to R284G or N154S/R284G, respectively, in human STING) are co-expressed under control of a CMV promoter, using T2A, and murine IL- 18 is expressed separately under control of an EF-la promoter. This exemplary construct also includes a WPRE for enhanced expression.

As described herein, the immunostimulatory bacteria can be used as vaccines in which the encoded antigen(s) is/are expressed under control of a bacterial promoter for transcription in the bacterium, but include regulatory sequences, such as an IRES, that prevents translation in the bacterium, but facilitates or permits translation in the eukaryotic host, such as a human. Nucleic acid constructs encoding such antigens can optionally encode additional immune system/response enhancing proteins, such as a STING protein, and/or a cytokine, and/or other combinations, which in effect, act as adjuvants. b. Dual/Multiple Promoter Constructs

Alternatively, the genetic payloads/therapeutic products can be expressed using dual or multiple promoter constructs, where each protein is expressed under the control of a separate promoter. Thus, plasmids encoding therapeutic products, such as immunomodulatory proteins, expressed in combinations, can contain multiple promoters, each controlling an individual intact ORF with proper stop codon processing (i.e., dual/multiple promoter constructs), or multiple proteins can be expressed in a single ORF through the use of 2A peptides (z.e., single promoter constructs); or the plasmid can contain a mixture of single and dual/multiple promoter constructs, to express three or more proteins, as described above.

3. Regulatory Elements a. Post-Transcriptional Regulatory Elements

In order to enhance expression of single and multiple therapeutic products/immunomodulatoiy proteins from a single plasmid, regulatory elements may be employed that enhance the transcription and translation of the protein(s) of interest. For example, the post-transcriptional regulatory element (PRE) of woodchuck hepatitis virus (WPRE), when inserted in the 3’ untranslated region of the ORF, can enhance expression levels several fold (see, e.g., Zufferey et al. (1999) J. Virol. 73(4):2886-2892). Similarly, other such elements, including, but not limited to, the Hepatitis B Virus PRE (HPRE), also can enhance expression. The combination of these can be used at the 3’ ends of multiple ORFs to improve expression of multiple proteins on a single plasmid.

The PREs WPRE and HPRE are hepadnaviral c/s-acting RNA elements that can increase the accumulation of cytoplasmic mRNA by promoting mRNA exportation from the nucleus, and can enhance post-transcriptional processing and stability. b. Polyadenylation Signal Sequences and Terminators

Other elements on the plasmid that can enhance protein expression include polyadenylation signal sequences and terminators. Polyadenylation is the post- transcriptional addition of a poly(A) tail to the 3’ end of an mRNA transcript, which is part of the process that produces mature mRNA for translation. Polyadenylation signal sequences are important for nuclear export, mRNA stability, and efficient translation. A terminator is a sequence that defines the end of a transcript, creating a free 3’ end, and initiates the release of the newly synthesized mRNA from the transcriptional machinery. The free 3’ end is then available for the addition of the poly(A) tail. Terminators are found downstream of the gene to be transcribed, and typically occur directly after any 3’ regulatory elements, such as the polyadenylation or poly(A) signal. Commonly used mammalian terminators in expression plasmids include the simian virus 40 (SV40), human growth hormone (hGH), bovine growth hormone (BGH or bGH), and rabbit beta-globin (rbGlob) poly(A) sequences, that include the sequence motif AAUAAA (SEQ ID NO:398), and promote both polyadenylation and termination.

When placed at the 3’ end of the ORF, sequences such as the simian virus 40 poly A (SV40pA) or the bovine growth hormone poly A (bGHpA) signals, result in several-fold increased expression both in vitro and in vivo (see, c.g., Powell etal. (2015) Discov. Med. 19(102):49-57). These and other such elements can further enhance the expression and translation of multiple therapeutic products, including immunomodulatory proteins, expressed from a single plasmid. c. Enhancers

Promoters and enhancers are found upstream of the multiple cloning site (MCS) in a plasmid, and cooperate to determine the rate of transcription. Enhancers are sequences that bind activator proteins, in order to loop the DNA, and bring a specific promoter to the initiation complex, thus increasing the rate of transcription. They can be adjacent to, or far from the promoter they influence, and include CMV, EF-la, SV40, and synthetic enhancers, or the MND promoter, which is a synthetic promoter that contains the U3 region of a modified MoMuLV (Moloney murine leukemia virus) LTR with myeloproliferative sarcoma virus enhancer. The immunostimulatory bacteria herein contain plasmids that can comprise enhancer(s) to enhance the expression of the therapeutic products/proteins encoded on the plasmids. d. Secretion Signals

A secretion signal, also known as a signal sequence or peptide, a leader sequence or peptide, or a localization signal or sequence, is a short peptide at the N- terminus of a newly synthesized protein that is to be secreted. Signal peptides promote a cell to translocate a protein, usually to the cellular membrane. The efficiency of protein secretion is strongly determined by the signal peptide. Thus, the immunostimulatory bacteria herein contain plasmids that can comprise a signal peptide/ secretion signal peptide, to facilitate and/or increase the expression or secretion of the encoded therapeutic product(s). e. Improving Bacterial Fitness

The plasmids in the immunostimulatory bacteria that encode the therapeutic products, include genes and regulatory elements that are provided for expression of bacterial genes, and also, for expression of complex polycistronic eukaryotic payloads. The switch between such evolutionarily divergent organisms introduces challenges for proper functioning in prokaryotes and eukaryotes. The bacteria are cultured in vitro, then administered to a eukaryotic subject, where the plasmids are delivered to cells, particularly to tumor-resident myeloid cells, in cancer subjects, where the payloads are expressed, processed and trafficked. Transcriptional leakiness from the eukaryotic promoter, such as the CMV promoter, in bacteria, combined with large eukaryotic genes and regulatory sequences, can result in reduced bacterial fitness that manifests as low injection stock viability, and reduced growth rate in broth culture.

4. Origin of Replication and Plasmid Copy Number

Plasmids are autonomously-replicating, extra-chromosomal circular doublestranded DNA molecules that are maintained within bacteria by means of a replication origin. Copy number influences the plasmid stability. High copy number generally results in greater stability of the plasmid when the random partitioning occurs at cell division. A high copy number of plasmids generally decreases the growth rate, thus possibly allowing for bacterial cells with few plasmids to dominate the culture, since they grow faster. This can be ameliorated by using gene attenuation and gene dosing strategies, that limit the expression of certain genes on the plasmid that can be toxic to the bacteria when present in high copy numbers. The origin of replication also determines the plasmid’s compatibility, i.e., its ability to replicate in conjunction with another plasmid within the same bacterial cell. Plasmids that utilize the same replication system cannot co-exist in the same bacterial cell; they are said to belong to the same compatibility group. The introduction of a new origin, in the form of a second plasmid from the same compatibility group, mimics the result of replication of the resident plasmid. Thus, any further replication is prevented until after the two plasmids have been segregated to different cells to create the correct pre-replication copy number.

Numerous bacterial origins of replication are known to those of skill in the art. The origin can be selected to achieve a desired copy number. Origins of replication contain sequences that are recognized as initiation sites of plasmid replication via DNA dependent DNA polymerases (see, e.g., del Solar et al. (1998) Microbiol. Mol. Biol. Rev. 62(2):434-464). Different origins of replication provide for varying plasmid copy levels within each cell, and can range from one to hundreds of copies per cell. Commonly used bacterial plasmid origins of replication include, but are not limited to, pMBl derived origins, which have very high copy derivatives, such as pUC, and lower copy derivatives, such as pBR322, as well as ColEl, pl5A, and pSClOl, and other origins, which have low copy numbers. Such origins are well-known to those of skill in the art. For example, the pUC19 origin results in a copy number of 500-700 copies per cell. The pBR322 origin has a known copy number of 15-20 copies per cell. These origins only vary by a single base pair. The ColEl origin copy number is 15-20, and derivatives, such as pBluescript, have copy numbers ranging from SOO- SOO. The pl5A origin that is in plasmid pACYC184, for example, results in a copy number of approximately 10. The pSClOl origins confer a copy number of approximately 5. Other low copy number vectors from which origins of replication can be obtained, include, for example, pWSK29, pWKS30, pWSK129, and pWKS130 (see, e.g., Wang et al. (1991) Gene 700: 195-199). Medium to low copy number is less than 150, or less than 100. Low copy number is less than 20, 25, or 30. Generally, less than medium copy number is less than 150 copies, and less than low copy number is less than about 25 or less than 25 copies, and generally, copy number refers to the average copies of plasmid per bacterium in a preparation. Those of skill in the art can identify plasmids with low, medium, or high copy numbers. For example, one method to determine experimentally if the copy number is high or low is to perform a miniprep. A high-copy plasmid should yield between 3-5 pg DNA per 1 ml LB culture; a low-copy plasmid will yield between 0.2-1 pg DNA per ml of LB culture. Sequences of bacterial plasmids, including identification of and sequence of the origin of replication, are well known (see, e.g., snapgene.com/resources/plasmid_files/basic_cloning_vectors/p BR322/). Exemplary origins of replication, and their plasmid copy numbers, are summarized in the table below.

High copy plasmids are selected for heterologous expression of proteins in vitro, because the gene dosage is increased relative to chromosomal genes, there are higher specific yields of protein, and for therapeutic bacteria, higher therapeutic dosages of encoded therapeutics. It is shown, herein, however, that for delivery of plasmids encoding therapeutic products (e.g, immunomodulatory proteins), such as by S. typhimurium, in some embodiments, a high copy plasmid might be advantageous.

The requirement for bacteria to maintain the high copy plasmids can be a problem if the expressed molecule is toxic to the organism. The metabolic requirements for maintaining these plasmids can come at a cost of replicative fitness in vivo. Optimal plasmid copy number for delivery of therapeutic products can depend on the mechanism of attenuation of the strain engineered to deliver the plasmid. If needed, the skilled person, in view of the disclosure herein, can select an appropriate copy number for a particular immunostimulatory species and strain of bacteria.

For use as vaccines and/or to deliver RNA (such as tnRNA) where the encoded products are expressed in the bacterium, the plasmid can be in higher or high copy number (greater than 150), to increase the amount of RNA or mRNA produced.

5. CpG Motifs and CpG Islands

Unmethylated cytidine-phosphate-guanosine (CpG) motifs are prevalent in bacterial, but not in vertebrate, genomic DNA. Pathogenic DNA and synthetic oligodeoxynucleotides (ODNs) containing CpG motifs activate host defense mechanisms, leading to innate and acquired immune responses. The unmethylated CpG motifs contain a central unmethylated CG dinucleotide plus flanking regions. In humans, four distinct classes of CpG ODNs have been identified, based on differences in structure, and in the nature of the immune response they induce. K-type ODNs (also referred to as B-type) contain from 1 to 5 CpG motifs, typically on a phosphorothioate backbone. D-type ODNs (also referred to as A-type) have a mixed phosphodiester/phosphorothioate backbone and have a single CpG motif, flanked by palindromic sequences that permit the formation of a stem-loop structure, as well as poly G motifs at the 3’ and 5’ ends. C-type ODNs have a phosphorothioate backbone and contain multiple palindromic CpG motifs that can form stem loop structures or dimers. P-Class CpG ODNs have a phosphorothioate backbone, and contain multiple CpG motifs with double palindromes that can form hairpins at their GC-rich 3’ ends (see, e.g., Scheiermann et al. (2014) Vaccine 32(48):6377-6389). For purposes herein, the CpGs are encoded in the plasmid DNA; they can be introduced as a motif, or in a gene.

Toll-like receptors (TLRs) are key receptors for sensing pathogen-associated molecular patterns (PAMPs) and activating innate immunity against pathogens (see, e.g., Akira et al. (2QQ ) Nat. Immunol. 2(8):675-680). TLR9 recognizes hypomethylated CpG motifs in the DNA of prokaryotes that do not occur naturally in mammalian DNA (see, e.g., McKelvey et al. (2011) J. Autoimmun. 36:76-86). Recognition of CpG motifs, upon phagocytosis of pathogens into endosomes in immune cell subsets, and activates innate and adaptive immunity.

Immunostimulatory bacteria, such as Salmonella species, such as S. typhimurium strains, carrying plasmids containing CpG islands or motifs, are provided herein. These bacteria can activate TLR9. As exemplified herein, bacterial plasmids that contain hypomethylated CpG islands can elicit innate and adaptive antitumor immune responses that, in combination with the therapeutic products encoded on the plasmid, such as immunostimulatory proteins and constitutively active variants of STING, IRF3, and other cytosolic DNA/RNA sensors, can have synergistic or enhanced anti-tumor activity. For example, the asd gene (see, e.g., SEQ ID NO:48) encodes a high frequency of hypomethylated CpG islands. CpG motifs can be included in combination with any of the therapeutic products, described or apparent from the description herein, in the immunostimulatory bacteria, to thereby enhance or improve anti -tumor immune responses in a treated subject.

Immunostimulatory CpGs can be included in the plasmids, by including a nucleic acid, typically from a bacterial gene, that encodes a gene product, and also by adding a nucleic acid that encodes CpG motifs. The plasmids herein can include CpG motifs. Exemplary CpG motifs are known (see, e.g., U.S. Patent Nos. 8,232,259, 8,426,375, and 8,241,844). These include, for example, synthetic immunostimulatory oligonucleotides, that are between 10 and 100, 10 and 20, 10 and 30, 10 and 40, 10 and 50, or 10 and 75, base pairs long, with the general formula: (CpG) _n , where n is the number of repeats. Generally, at least one or two repeats are used; non-CG bases can be interspersed. Those of skill in the art are very familiar with the general use of CpG motifs for inducing an immune response by modulating TLRs, particularly TLR9.

6. Plasmid Maintenance/Selection Components

The maintenance of plasmids in laboratory settings is usually ensured by the inclusion of an antibiotic resistance gene on the plasmid, and the use of antibiotics in the growth media. As described above, the use of an asd deletion mutant, complemented with a functional asd gene on the plasmid, allows for plasmid selection in vitro without the use of antibiotics, and allows for plasmid maintenance in vivo. The asd gene complementation system provides for such selection/maintenance (see, e.g., Galan et al. (1990) Gene 94(l):29-35). The use of the asd gene complementation system to maintain plasmids in the tumor microenvironment increases the potency of S. typhimurium and other immunostimulatory bacterial strains, engineered to deliver plasmids encoding therapeutic products, such as immunostimulatory proteins, constitutively active cytosolic DNA/RNA sensors, antibodies, antibody fragments, or other such products as discussed herein.

7. DNA Nuclear Targeting Sequences

DNA nuclear targeting sequences (DTS), such as the SV40 DTS, mediate the translocation of DNA sequences through the nuclear pore complex. The mechanism of this transport is reported to be dependent on the binding of DNA binding proteins that contain nuclear localization sequences. The inclusion of a DTS on a plasmid to increase nuclear transport and expression has been demonstrated (see, e.g., Dean, D.A. et al. (1999) Exp. Cell Res. 253(2):713-722), and has been used to increase gene expression from plasmids delivered by S. typhimurium (see, e.g., Kong et al. (2012) Proc. Natl. Acad. Sci. U.S.A. 109(47): 19414-19419).

Rho-independent or class I transcriptional terminators, such as the T1 terminator of the rrnB gene of E. coli, contain sequences of DNA that form secondary structures that cause dissociation of the transcription elongation complex. Transcriptional terminators are included in the plasmid in order to prevent expression of heterologous proteins by the S. typhimurium transcriptional machinery. This ensures that expression of the therapeutic products is confined to the host cell transcriptional machinery.

Plasmids used for transformation of Salmonella, such as S. typhimurium, as a cancer therapy described herein, contain all or some of the following attributes: 1) one or more constitutive promoters for heterologous expression of proteins; 2) one or more human immunomodulatory expression cassettes; 3) a bacterial origin of replication and optimized plasmid copy number; 4) immunostimulatory CpG islands; 5) an asd gene selectable marker for plasmid maintenance and selection; 6) DNA nuclear targeting sequences; and 7) transcriptional terminators.

G. EXEMPLARY BACTERIAL STRAINS AND MECHANISM OF ACTION FOR USE AS VACCINES AND THERAPEUTICS

As described throughout, provided are immunostimulatory bacteria that contain genome modifications that improve their properties for use as anti-cancer therapeutics, as vaccines against pathogens, and as delivery vehicles for RNA, which can be used for as therapeutic products, or, that, can be translated to provide immunostimulatory products, antigens, and combinations thereof for the various uses. The immunostimulatory bacteria include genomic modifications discussed throughout that improve their properties for use as anti-cancer therapeutics, for vaccines, and RNA delivery platforms, and also that reduce the inflammatory properties of bacteria, also include payloads, including combinations of payload products, that enhance or act with the properties of the bacteria for these applications. Payloads and genomic modifications are discussed and described throughout the disclosure herein, and are exemplified in the working examples below. The following is a discussion of exemplary bacteria, combinations of genomic modifications and payloads, and particularly advantageous for their use as vaccines. Disclosure throughout describes properties, payloads, and modifications for use as cancer therapeutics, and as therapeutics that are generally useful for appropriate immune stimulation or modulation for use for treating cancer and other diseases and conditions.

It is understood that the skilled artisan can effect similar genomic modifications in other bacterial species and strains, such as E. coli, and Listeria, to achieve similar effects and results. Additionally, as discussed above won-Sa! m one IE species can be modified to encode a rck gene, such the rck gene from Salmonella, to increase resistance to complement.

For vaccination, the use and general mechanism of action for the bacteria provided herein includes administration, particularly direct tissue administration, such as by intramuscular injection, inhalation, intradermal administration, vaginal administration, and other such routes, to generate an in situ immune response to the antigen or product, which can be for prevention (immunization to develop an immmunoprotective response) or treatment, for a pathogen or for cancer. The bacteria induce durable adaptive antiviral immunity. The bacteria are suitably formulated for the selected route. This can include formulation, for example, as an aerosol, tablet, emulsion, powder, as appropriate. The so-formulated administered bacteria are taken up into tissue-resident phagocytes, by phagocytosis. For these purposes, the bacteria are modified so that they do no replicate in vivo, but deliver their payloads into the cells, such as tissue-resident phagocytic antigen-presenting cells (APCs). The encoded products include, for example antigens for presentation, STING to induce type I IFN. The encoded products can go into the lymph nodes resulting in antibodies via CD4+ and CD8+ priming, and the APCs present for priming and activation of antigenspecific CD8+ cells, which stimulate IFN-gamma, and also participate in direct killing of virally infected cells; macrophages phagocytose apoptotic infected cells, thereby spreading epitopes, which results in further T-cell activation

1. Exemplary Immunostimulatory Bacteria — In Situ Cancer Vaccination Mechanism of Action (MOA)

Among the immunostimulatory bacteria provided herein are immunostimulatory bacteria that are designed to accumulate in tumor-resident myeloid cells, such as macrophage, and other phagocytic cells, such as when subjects are vaccinated with bacteria designed as vaccines. Bacteria provided herein contain a plasmid that encodes immunostimulatory proteins and combinations thereof that induce an anti-viral/anti-tumor immune response. The genomic modifications include the elimination of flagella, alterations of the LPS, and elimination of curli fimbriae. These or modifications in bacterial strains with similar effects, reduce TLR2/4/5 responses/signaling, and result in accumulation in the tumor microenvironment and myeloid cells. Other modifications include auxotrophies, such as asd~ or thyA' that render the bacteria unable to grow in vivo, and other auxotrophies, such as including purl', that render the bacteria auxotrophic for adenosine, and other modifications that increase their accumulation in the tumor microenvironment, which can have elevated adenosine, which is immunosuppressive, and genomic modifications to eliminate or inactivate asparaginase, which can inhibit T-cells. The combination of some or all of these bacterial modifications results in bacteria that accumulate in the macrophage in the tumor microenvironment and tumors. Exemplary of such immunostimulatory bacteria is the strain designated \M6Aasdl MAAjIApagP/AansB/AcsgD/F-Apurl strain. The strain can be used, for example, for in situ cancer antigen vaccination, where the strain contains a plasmid expressing either STING, eSTING (referring to engineered STING, such as those provided herein, particularly those that constitutively induce type I interferon), IL- 15 (or IL- 15 receptor complex) + eSTING, and other type I IFN producing combinations of payloads, expressed under control of a eukaryotic promoter, such as the CMV promoter. The plasmid additionally contains sequences encoding the open reading frames of full-length tumor antigens or peptide antigen fragments. The route of administration is IV, whereby the strain traffics to the tumor and is taken up by tumor-resident macrophages, resulting in plasmid transfer to the nucleus and expression of payloads. In the presence of type I IFN, the macrophages process and present the cancer antigens for CD8 ⁺ T-cell priming within the tumor, as well as trafficking to the lymph nodes to prime CD4 ⁺ and CD8 ⁺ T-cells. Primed and activated CD8 ⁺ T-cells eliminate tumor cells bearing the primed antigen, induce long-lived tissue-resident memory CD8+ T-cells, as well as circulating as memory T-cells after the tumor is cleared to prevent tumor recurrence.

2. Exemplary immunostimulatory bacteria for Peripheral Cancer Vaccination and Mechanism of Action

The Y S I 646Ai.s<7/ AFLG/ ApagP/AansB/AcsgD/F-Apurl/AthyA strain containing an additional thymidine auxotrophy is used for peripheral cancer antigen vaccination and tumor treatment. The strain cannot proliferate in the absence of thymidine supplementation, and, upon administration, as by intramuscular injection into an extremity, it is phagocytosed by tissue-resident macrophages, such as following IM injection in the arm. The strain contains a plasmid that encodes STING, eSTING, IL- 15 + eSTING, or other type I IFN producing combination, as well as sequences of full-length tumor antigens or peptides, from either a bacterial promoter and IRES sequence or a eukaryotic promoter. For the bacterial promoter containing strains, the bacteria will then transcribe the RNA of these encoded payloads, and deliver them upon phagocytosis and destruction by tissue-resident macrophages. For the eukaryotic promoter containing strains, the phagocyte transcribes the RNA. The resulting mRNA the is translated into proteins, and the encoded antigens are presented by the macrophages in the presence of type I IFN for CD8 ⁺ T-cell priming. The cancer antigen-primed CD8 ⁺ T-cells circulate and eliminate cancerous cells, such as those that remain after surgical tumor resection. Strains such as this also are used as a prophylactic vaccine to reduce the risk of developing a particular cancer, such as for vaccination against familial genetic mutations that could lead to tumor formation. 3. Exemplary Immunostimulatory Bacteria for Pathogen Vaccination and Mechanism of Action (MOA)

Also provided are bacterial strains that include additional auxotrophy, such as thyA~ strains. Exemplary of such include YS1646Aa%Z/AFLG/A/?^g /Aaw _J 8/AcsgD/F-A/2Mr//A/A 4 strain, which has an additional thymidine auxotrophy. Such strains cannot proliferate in the absence of thymidine supplementation. These strains can be used for vaccination against pathogens. They are administered by route that infection by the particular pathogen occurs naturally. Following administration, such as by intramuscular, intranasal, inhalation, or oral delivery, or combinations thereof, depending on the site or sites where the infection naturally occurs, they are cleared from tissue-resident macrophages, leaving plasmids. These strains contain a plasmid encoding an immunostimulatory protein or combination thereof, such as STING, engineered STING (eSTING), IL-15 + eSTING, or other type I IFN producing proteins or combinations of proteins, and also pathogen sequences of antigens and/or peptides, under transcriptional control of a bacterial promoter, but including eukaryotic sequences, such as an IRES, for translation by eukaryotic ribosomes. The bacteria produce RNA encoding these payloads, and, after administration, upon phagocytosis and destruction by tissue-resident macrophages, deliver the encoded mRNA. The mRNA is translated into proteins by the eukaryotic host cells, and the antigens are presented by the macrophages in the presence of type I IFN for CD8 ⁺ T-cell priming. The type I IFN can be stimulated by the encoded immunostimulatory protein, such as a STING or modified STING. Additionally, the bacteria can provide the type I IFN and pathogen antigens as proteins using a plasmid with a bacterial promoter and sequences for bacterial transcription and translation. The macrophages then prime CD8 ⁺ T-cells in the tissue, as well as traffic to the lymph nodes to prime CD4 ⁺ and CD8 ⁺ T-cells. Primed and activated CD8 ⁺ T-cells then circulate as patrolling memory T-cells, as well as traffic back to the tissue of origin to remain as tissue-resident memory CD8 ⁺ T-cells. The primed CD4 ⁺ T-cells induce B-cell activation to produce high titers of durable neutralizing antibodies. This vaccination strategy mimics the highly successful smallpox, polio, measles, chicken pox, and other live-attenuated vaccines that are administered in the tissues where disease would naturally occur.

4. Exemplary Immunostimulatory Bacteria for Treatment of Cancer As discussed throughout the disclosure herein, bacteria provide herein can be administered, such as systemically, for treatment of cancer. The bacteria, such as strains that have phenotypes that are like YS \(A(>&asdl AFLG/ kpagPI '/ \csgDI or YS1646A<2.5’c//AFLG/ApagP/Aart.s'B/Ac.sgD/F-A7?Mr/, or other such strains that have penta-acylated LPS, lack curli fimbriae, lack flagella, and are adenosine auxotrophs, and other optional genome modifications, target tumor-resident myeloid cells, and are taken up phagocytic cells. These bacteria are well-tolerated and non-toxic, even at high doses.

These bacteria, which are phagocytosed by phagocytic cells, encode therapeutic proteins, particularly proteins used for anti-cancer treatment. Exemplary of the encoded payloads are the combination of a cytokinie, such as IL-15/IL-15R alpha chain complex, and a modified cytosolic DNA/RNA sensor protein that induce type I interferon (IFN), such as the modified STING proteins (eSTING) that constitutively induce type I interferon (IFN) in phagocytic cells, such as proliferating macrophage. Other payloads of interest include a type I interferon (IFN), such as and IFNa and/or IFNb, and also combinations with other proteins, such as a tumor- associated antigens. The Examples below exemplify the anti-tumor activities that result upon administration of these bacteria.

It is shown herein, for example, that the bacteria delivering combination of IL- 15/IL-15R alpha chain complex and eSTING (and similar combinations) synergistically activate human dendritic cells to potently stimulate antigen-specific CD8+ T-cell responses. These bacteria are shown to result in a higher cure rate in rodent models, such as the difficult-to-treat EMT6 metastatic orthotopic TNBC model. Intravenous administration of these bacteria demonstrates significant T-cell infiltration in T-cell excluded tumors, such as the metastatic EMT6 TNBC breast cancer, including demonstrating cures of metastatic disease and protection from orthotopic tumor re-challenge in a CD8+ T-cell dependent manner. As shown in the Examples, these bacteria demonstrate tumor colonization and T-cell infiltration in a breast cancer immune desert model. As described above and demonstrated in the examples, these bacteria result in a heretofore not observed advantageous hybrid M1//M2 phenotype.

As shown in the examples, immuno stimulatory bacteria provided herein render checkpoint refractory tumors susceptible to anti-checkpoint inhibitors, such as anti-PD-Ll and anti-PD-1 therapies.

- 307 -

RECTIFIED SHEET (RULE 91) ISA/EP H. PHARMACEUTICAL PRODUCTION, COMPOSITIONS, AND FORMULATIONS

Provided herein are methods for manufacturing, and pharmaceutical compositions and formulations, containing any of the immunostimulatory bacteria provided herein and pharmaceutically acceptable excipients or additives. The pharmaceutical compositions can be used in the treatment of diseases, such as hyperproliferative diseases or conditions, such as a tumor or cancer. The immunostimulatory bacteria can be administered as a single agent therapy, or can be administered in a combination therapy with a further agent(s) or treatment(s). Combination therapy includes combining therapy with the immunostimulatory bacteria and/or other delivery vehicles provided herein, with any other anti -cancer therapy or treatment, including, but not limited to, immunotherapies, such as CAR-T therapy and checkpoint inhibitors, radiation, surgery, chemotherapeutic agents, such as nucleoside analogs and platinum compounds, and cellular therapies. The compositions can be formulated for single dosage administration, or for multiple dosage administration. The agents can be formulated for direct administration. The compositions can be provided as a liquid or dried formulation.

1. Manufacturing a. Cell Bank Manufacturing

As the active ingredient of the immunotherapeutic described herein is composed of engineered self-replicating bacteria, the selected composition will be expanded into a series of cell banks that will be maintained for long-term storage and as the starting material for manufacturing of the drug substance. Cell banks are produced under current good manufacturing practices (cGMP) in an appropriate manufacturing facility per the Code of Federal Regulations (CFR) 21 part 211, or other relevant regulatory authority. As the active agent of the immunotherapeutic is a live bacterium, the products described herein are, by definition, non-sterile and cannot be terminally sterilized. Care must be taken to ensure that aseptic procedures are used throughout the manufacturing process to prevent contamination. As such, all raw materials and solutions must be sterilized prior to use in the manufacturing process.

A master cell bank (MCB) is produced by sequential serial single colony isolation of the selected bacterial strain, to ensure no contaminants are present in the starting material. A sterile culture vessel containing sterile media (can be complex media, e.g., LB or MSB, or defined media, e.g., M9 supplemented with appropriate nutrients) is inoculated with a single well-isolated bacterial colony and the bacteria are allowed to replicate, e.g., by incubation at 37 °C with shaking. The bacteria are then prepared for cryopreservation by suspension in a solution containing a ciyoprotective agent or agents.

Examples of cryoprotective agents include: proteins, such as human or bovine serum albumin, gelatin, and immunoglobulins; carbohydrates, including monosaccharides (e.g., galactose, D-mannose, sorbose, etc.) and their non-reducing derivatives (e.g., methylglucoside), disaccharides (e.g., trehalose, sucrose, and others), cyclodextrins, and polysaccharides (e.g., raffinose, maltodextrins, dextrans, etc. amino-acids (e.g., glutamate, glycine, alanine, arginine or histidine, tryptophan, tyrosine, leucine, phenylalanine, etc.) methylamines, such as betaine; polyols, such as trihydric or higher sugar alcohols, e.g., glycerin, erythritol, glycerol, arabitol, xylitol, sorbitol, and mannitol; propylene glycol; polyethylene glycol; surfactants, e.g., Pluronic®; or organo-sulfur compounds, such as dimethyl sulfoxide (DMSO), and combinations thereof. Cry opreservation solutions can include one or more cryoprotective agents in a solution that also can contain salts (e.g., sodium chloride, potassium chloride, magnesium sulfate), and/or buffering agents, such as sodium phosphate, tris(hydroxymethyl)aminomethane (TRIS), 4-(2 -hy droxy ethyl)- 1- piperazineethanesulfonic acid (HEPES), and other such buffering agents known to those of skill in the art.

Suspension of the bacteria in cryopreservation solution can be achieved either by addition of a concentrated cryoprotective agent or agents to the culture material to achieve a final concentration that preserves viability of the bacteria during the freezing and thawing process (e.g., 0.5% to 20% final concentration of glycerol), or by harvesting the bacteria (e.g., by centrifugation) and suspending in a cryopreservative solution containing the appropriate final concentration of cryoprotective agent(s). The suspension of bacteria in cryopreservation solution is then filled into appropriate sterile vials (plastic or glass) with a container closure system that is capable of maintaining closure integrity under frozen conditions (e.g., butyl stoppers and crimp seals). The vials of master cell bank are then frozen (either slowly by means of a controlled rate freezer, or quickly by means of placing directly into a freezer). The MCB is then stored frozen at a temperature that preserves longterm viability (e.g., at or below -60 °C). Thawed master cell bank material is thoroughly characterized to ensure identity, purity, and activity per regulation by the appropriate authorities.

Working cell banks (WCBs) are produced much the same way as the master cell bank, but the starting material is derived from the MCB. MCB material can be directly transferred into a fermentation vessel containing sterile media and expanded as above. The bacteria are then suspended in a cry opreservation solution, filled into containers, sealed, and frozen at or below -20 °C. Multiple WCBs can be produced from MCB material, and WCB material can be used to make additional cell banks (e.g., a manufacturer’s working cell bank (MWCB)). WCBs are stored frozen, and are characterized to ensure identity, purity, and activity. WCB material is typically the starting material used in the production of the drug substance of biologies such as engineered bacteria. b. Drug Substance Manufacturing

Drug substance is manufactured using aseptic processes under cGMP, as described above. Working cell bank material is typically used as starting material for manufacturing of drug substance under cGMP, however, other cell banks can be used (e.g., MCB or MWCB). Aseptic processing is used for production of all cell therapies including bacterial cell-based therapies. The bacteria from the cell bank are expanded by fermentation; this can be achieved by production of a pre-culture (e.g., in a shake flask), or by direct inoculation of a fermenter. Fermentation is accomplished in a sterile bioreactor or flask that can be single-use disposable, or re-usable. Bacteria are harvested by concentration (e.g., by centrifugation, continuous centrifugation, or tangential flow filtration). Concentrated bacteria are purified from media components and bacterial metabolites by exchange of the media with buffer (e.g., by diafiltration). The bulk drug product is formulated and preserved as an intermediate (e.g., by freezing or drying), or is processed directly into a drug product. Drug substance is tested for identity, strength, purity, potency, and quality. c. Drug Product Manufacturing

Drug product is defined as the final formulation of the active substance contained in its final container. Drug product is manufactured using aseptic processes under cGMP. Drug product is produced from drug substance. Drug substance is thawed or reconstituted if necessary, then formulated at the appropriate target strength. Because the active component of the drug product is live, engineered bacteria, the strength is determined by the number of colony forming units (CFUs) contained within the suspension. The bulk product is diluted in a final formulation appropriate for storage and use, as described below. Containers are filled and sealed with a container closure system, and the drug product is labeled. The drug product is stored at an appropriate temperature to preserve stability and is tested for identity, strength, purity, potency, and quality, and released for human use if it meets specified acceptance criteria.

2. Compositions

Pharmaceutically acceptable compositions are prepared in view of approvals for a regulatory agency or other agency, and/or prepared in accordance with generally recognized pharmacopeia for use in animals and in humans. The compositions can be prepared as solutions, suspensions, powders, or sustained release formulations. Typically, the compounds are formulated into pharmaceutical compositions using techniques and procedures well-known in the art (see, e.g, Ansel, Introduction to Pharmaceutical Dosage Forms , Fourth Edition, 1985, page 126). The formulation should suit the mode of administration.

Compositions can be formulated for administration by any route known to those of skill in the art, including intramuscular, intravenous, intradermal, intralesional, intraperitoneal, subcutaneous, intratumoral, epidural, nasal, oral, vaginal, rectal, topical, local, otic, inhalational, buccal (e.g, sublingual), and transdermal administration, or by any suitable route. Other modes of administration also are contemplated. Administration can be local, topical, or systemic, depending upon the locus of treatment. Local administration to an area in need of treatment can be achieved by, for example, but not limited to, local infusion during surgery, topical application, e.g., in conjunction with a wound dressing after surgery, by injection, by means of a catheter, by means of a suppository, or by means of an implant. Compositions also can be administered with other biologically active agents, either sequentially, intermittently, or in the same composition. Administration also can include controlled release systems, including controlled release formulations and device controlled release, such as by means of a pump.

The most suitable route in any given case depends on a variety of factors, such as the nature of the disease, the progress of the disease, the severity of the disease, and the particular composition which is used. Pharmaceutical compositions can be formulated in dosage forms appropriate for each route of administration. In particular, the compositions can be formulated into any suitable pharmaceutical preparations for systemic, local, intraperitoneal, oral, or direct administration. For example, the compositions can be formulated for administration subcutaneously, intramuscularly, intratumorally, intravenously, or intradermally. Administration methods can be employed to decrease the exposure of the active agent to degradative processes, such as immunological intervention via antigenic and immunogenic responses. Examples of such methods include local administration at the site of treatment, or continuous infusion.

The immunostimulatory bacteria can be formulated into suitable pharmaceutical preparations, such as solutions, suspensions, tablets, dispersible tablets, pills, capsules, powders, sustained release formulations, or elixirs, for oral administrations, as well as transdermal patch preparations, and dry powder inhalers. Typically, the compounds are formulated into pharmaceutical compositions using techniques and procedures well-known in the art (see, e.g, Ansel, Introduction to Pharmaceutical Dosage Forms, Fourth Edition, 1985, page 126). Generally, the mode of formulation is a function of the route of administration. The compositions can be formulated in dried (lyophilized or other forms of vitrification) or liquid form. Where the compositions are provided in dried form, they can be reconstituted just prior to use by addition of an appropriate buffer, for example, a sterile saline solution.

3. Formulations a. Liquids, Injectables, Emulsions

The formulation generally is made to suit the route of administration. Parenteral administration, generally characterized by injection or infusion, either subcutaneously, intramuscularly, intratumorally, intravenously, or intradermally is contemplated herein. Preparations of bacteria for parenteral administration include suspensions ready for injection (direct administration), frozen suspensions that are thawed prior to use, dry soluble products, such as lyophilized powders, ready to be combined with a resuspension solution just prior to use, and emulsions. Dried thermostable formulations, such as lyophilized formulations, can be used for storage of unit doses for later use.

The pharmaceutical preparation can be in a frozen liquid form, for example, a suspension. If provided in frozen liquid form, the drug product can be provided as a concentrated preparation to be thawed and diluted to a therapeutically effective concentration before use. The pharmaceutical preparations also can be provided in a dosage form that does not require thawing or dilution for use. Such liquid preparations can be prepared by conventional means with pharmaceutically acceptable additives, as appropriate, such as suspending agents (e.g., sorbitol, cellulose derivatives, or hydrogenated edible fats); emulsifying agents (e.g., lecithin or acacia); non-aqueous vehicles (e.g., almond oil, oily esters, or fractionated vegetable oils); and preservatives suitable for use with microbial therapeutics. The pharmaceutical preparations can be presented in dried form, such as lyophilized or spray-dried, for reconstitution with water or other sterile suitable vehicle before use.

Suitable excipients are, for example, water, saline, dextrose, or glycerol. The solutions can be either aqueous or non-aqueous. If administered intravenously, suitable carriers include physiological saline or phosphate buffered saline (PBS), and other buffered solutions used for intravenous hydration. For intratumoral administration, solutions containing thickening agents, such as glucose, polyethylene glycol, and polypropylene glycol, oil emulsions, and mixtures thereof can be appropriate to maintain localization of the inj ectant.

Pharmaceutical compositions can include carriers or other excipients. For example, pharmaceutical compositions provided herein can contain any one or more of a diluents(s), adjuvant(s), antiadherent(s), binder(s), coating(s), filler(s), flavor(s), color(s), lubricant(s), glidant(s), preservative(s), detergent(s), or sorbent(s), and a combination thereof, or a vehicle with which a modified therapeutic bacteria is administered. For example, pharmaceutically acceptable carriers or excipients used in parenteral preparations include aqueous vehicles, non-aqueous vehicles, isotonic agents, buffers, antioxidants, local anesthetics, suspending and dispersing agents, emulsifying agents, sequestering or chelating agents, and other pharmaceutically acceptable substances. Formulations, including liquid preparations, can be prepared by conventional means with pharmaceutically acceptable additives or excipients.

Pharmaceutical compositions can include carriers, such as a diluent, adjuvant, excipient, or vehicle, with which the compositions are administered. Examples of suitable pharmaceutical carriers are described in “Remington’s Pharmaceutical Sciences” by E. W. Martin. Such compositions will contain a therapeutically effective amount of the compound or agent, generally in purified form or partially purified form, together with a suitable amount of carrier, so as to provide the form for proper administration to the patient. Such pharmaceutical carriers can be sterile liquids, such as water and oils, including those of petroleum, animal, vegetable, or synthetic origin, such as peanut oil, soybean oil, mineral oil, and sesame oil. Water is a typical carrier. Saline solutions and aqueous dextrose and glycerol solutions also can be employed as liquid carriers, particularly for injectable solutions. Compositions can contain along with an active ingredient: a diluent, such as lactose, sucrose, dicalcium phosphate, or carboxymethylcellulose; a lubricant, such as magnesium stearate, calcium stearate, and talc; and a binder, such as starch, natural gums, such as gum acacia, gelatin, glucose, molasses, polyvinylpyrrolidine, celluloses and derivatives thereof, povidone, crospovidone, and other such binders known to those of skill in the art. Suitable pharmaceutical excipients include starch, glucose, lactose, sucrose, gelatin, malt, rice, flour, chalk, silica gel, sodium stearate, glycerol monostearate, talc, sodium chloride, dried skim milk, glycerol, propylene, glycol, water, and ethanol. For example, suitable excipients are, for example, water, saline, dextrose, glycerol, or ethanol. A composition, if desired, also can contain other minor amounts of non-toxic auxiliary substances, such as wetting or emulsifying agents, pH buffering agents, stabilizers, solubility enhancers, and other such agents, such as, for example, sodium acetate, sorbitan monolaurate, triethanolamine oleate, and cyclodextrins.

Pharmaceutically acceptable carriers used in parenteral preparations include aqueous vehicles, non-aqueous vehicles, antimicrobial agents, isotonic agents, buffers, antioxidants, local anesthetics, suspending and dispersing agents, emulsifying agents, sequestering or chelating agents and other pharmaceutically acceptable substances. Examples of aqueous vehicles include Sodium Chloride Injection, Ringer’s Injection, Isotonic Dextrose Injection, Sterile Water Injection, and Dextrose and Lactated Ringer’s Injection. Non-aqueous parenteral vehicles include fixed oils of vegetable origin, cottonseed oil, corn oil, sesame oil, and peanut oil. Isotonic agents include sodium chloride and dextrose. Buffers include phosphate and citrate. Antioxidants include sodium bisulfate. Local anesthetics include procaine hydrochloride. Suspending and dispersing agents include sodium carboxymethylcellulose, hydroxypropyl methylcellulose, and polyvinylpyrrolidone. Emulsifying agents include, for example, polysorbates, such Polysorbate 80 (TWEEN 80). Sequestering or chelating agents of metal ions, such as EDTA, can be included. Pharmaceutical carriers also include polyethylene glycol, and propylene glycol for water miscible vehicles, and sodium hydroxide, hydrochloric acid, citric acid, or lactic acid, for pH adjustment. Non-anti-microbial preservatives can be included. The pharmaceutical compositions also can contain other minor amounts of non-toxic auxiliary substances, such as wetting or emulsifying agents, pH buffering agents, stabilizers, solubility enhancers, and other such agents, such as, for example, sodium acetate, sorbitan monolaurate, triethanolamine oleate, and cyclodextrins. Implantation of a slow-release or sustained-release system, such that a constant level of dosage is maintained (see, e.g., U.S. Patent No. 3,710,795), also is contemplated herein. The percentage of active compound contained in such parenteral compositions is highly dependent on the specific nature thereof, as well as the activity of the compound and the needs of the subj ect. b. Dried Thermostable Formulations

The bacteria can be dried. Dried thermostable formulations, such as lyophilized or spray dried powders and vitrified glass, can be reconstituted for administration as solutions, emulsions and other mixtures. The dried thermostable formulations can be prepared from any of the liquid formulations, such as the suspensions, described above. The pharmaceutical preparations can be presented in lyophilized or vitrified form, for reconstitution with water or other suitable vehicle, before use.

The thermostable formulation is prepared for administration by reconstituting the dried compound with a sterile solution. The solution can contain an excipient which improves the stability or other pharmacological attribute of the active substance or reconstituted solution, prepared from the powder. The thermostable formulation is prepared by dissolving an excipient, such as dextrose, sorbitol, fructose, com syrup, xylitol, glycerin, glucose, sucrose, or other suitable agent, in a suitable buffer, such as citrate, sodium or potassium phosphate, or other such buffer known to those of skill in the art. Then, the drug substance is added to the resulting mixture, and stirred until it is mixed. The resulting mixture is apportioned into vials for drying. Each vial will contain a single dosage, containing IxlO ⁵ to IxlO ¹¹ CFUs per vial. After drying, the product vial is sealed with a container closure system that prevents moisture or contaminants from entering the sealed vial. The dried product can be stored under appropriate conditions, such as at -20 °C, 4 °C, or room temperature. Reconstitution of this dried formulation with water or a buffer solution provides a formulation for use in parenteral administration. The precise amount depends upon the indication treated and selected compound. Such amount can be empirically determined. 4. Compositions for Other Routes of Administration

Depending upon the condition treated, other routes of administration in addition to parenteral, such as topical application, transdermal patches, and oral and rectal administration, also are contemplated herein. The suspensions and powders described above can be administered orally, or can be reconstituted for oral administration. Pharmaceutical dosage forms for rectal administration are rectal suppositories, capsules, and tablets and gel capsules for systemic effect. Rectal suppositories include solid bodies for insertion into the rectum which melt or soften at body temperature, releasing one or more pharmacologically or therapeutically active ingredients. Pharmaceutically acceptable substances in rectal suppositories are bases or vehicles and agents to raise the melting point. Examples of bases include cocoa butter (theobroma oil), glycerin-gelatin, CARBOWAX® (polyethylene glycol), and appropriate mixtures of mono-, di-, and triglycerides of fatty acids. Combinations of the various bases can be used. Agents to raise the melting point of suppositories include spermaceti and wax. Rectal suppositories can be prepared either by the compressed method, or by molding. The typical weight of a rectal suppository is about 2 to 3 grams. Tablets and capsules for rectal administration are manufactured using the same pharmaceutically acceptable substance and by the same methods as for formulations for oral administration. Formulations suitable for rectal administration can be provided as unit dose suppositories. These can be prepared by admixing the drug substance with one or more conventional solid carriers, for example, cocoa butter, and then shaping the resulting mixture.

For oral administration, pharmaceutical compositions can take the form of, for example, tablets or capsules prepared by conventional means with pharmaceutically acceptable excipients, such as binding agents (e.g, pregelatinized maize starch, polyvinyl pyrrolidone, or hydroxypropyl methylcellulose); fillers (e.g., lactose, microcrystalline cellulose, or calcium hydrogen phosphate); lubricants (e.g., magnesium stearate, talc, or silica); disintegrants (e.g, potato starch or sodium starch glycolate); or wetting agents (e.g., sodium lauryl sulfate). The tablets can be coated by methods well-known in the art.

Formulations suitable for buccal (sublingual) administration include, for example, lozenges containing the active compound in a flavored base, usually sucrose and acacia or tragacanth; and pastilles containing the compound in an inert base, such as gelatin and glycerin, or sucrose and acacia. Topical mixtures are prepared as described for local and systemic administration. The resulting mixtures can be solutions, suspensions, emulsions or the like, and are formulated as creams, gels, ointments, emulsions, solutions, elixirs, lotions, suspensions, tinctures, pastes, foams, aerosols, irrigations, sprays, suppositories, bandages, dermal patches, or any other formulations suitable for topical administration.

The compositions can be formulated as aerosols for topical application, such as by inhalation (see, e.g., U.S. Patent Nos. 4,044,126, 4,414,209, and 4,364,923, which describe aerosols for the delivery of a steroid useful for treatment of lung diseases). These formulations, for administration to the respiratory tract, can be in the form of an aerosol or solution for a nebulizer, or as a microfine powder for insufflation, alone or in combination with an inert carrier such as lactose. In such a case, the particles of the formulation will typically have diameters of less than 50 microns, or less than 10 microns.

The compounds can be formulated for local or topical application, such as for topical application to the skin and mucous membranes, such as in the eye, in the form of gels, creams, and lotions, and for application to the eye, or for intraci sternal or intraspinal application. Topical administration is contemplated for transdermal delivery, and also for administration to the eyes or mucosa, or for inhalation therapies. Nasal solutions of the active compound alone, or in combination with other pharmaceutically acceptable excipients, also can be administered.

Formulations suitable for transdermal administration are provided. They can be provided in any suitable format, such as discrete patches adapted to remain in intimate contact with the epidermis of the recipient for a prolonged period of time. Such patches contain the active compound in an optionally buffered aqueous solution of, for example, 0.1 to 0.2 M concentration, with respect to the active compound. Formulations suitable for transdermal administration also can be delivered by iontophoresis (see, e. ., Tyle, P. (1986) Pharmaceutical Research 3(6):318-326), and typically take the form of an optionally buffered aqueous solution of the active compound.

Pharmaceutical compositions also can be administered by controlled release formulations and/or delivery devices (see e.g., U.S. Patent Nos. 3,536,809; 3,598,123; 3,630,200; 3,845,770; 3,916,899; 4,008,719; 4,769,027; 5,059,595; 5,073,543; 5,120,548; 5,591,767; 5,639,476; 5,674,533; and 5,733,566). 5. Dosages and Administration

The compositions can be formulated as pharmaceutical compositions for single dosage or multiple dosage administration. The immunostimulatory bacteria can be included in an amount sufficient to exert a therapeutically useful effect in the absence of undesirable side effects on the patient treated. For example, the concentration of the pharmaceutically active compound is adjusted so that an injection provides an effective amount to produce the desired pharmacological effect. The therapeutically effective concentration can be determined empirically by testing the immunostimulatory bacteria in known in vitro and in vivo systems, such as by using the assays described herein or known in the art. For example, standard clinical techniques can be employed. In vitro assays and animal models can be employed to help identify optimal dosage ranges. The precise dose, which can be determined empirically, can depend on the age, weight, body surface area, and condition of the patient or animal, the particular immunostimulatory bacteria administered, the route of administration, the type of disease to be treated, and the seriousness of the disease.

Hence, it is understood that the precise dosage and duration of treatment is a function of the disease being treated, and can be determined empirically using known testing protocols, or by extrapolation from in vivo or in vitro test data. Concentrations and dosage values also can vary with the severity of the condition to be alleviated. It is to be further understood that, for any particular subject, specific dosage regimens should be adjusted over time according to the individual need and the professional judgment of the person administering or supervising the administration of the compositions, and that the concentration ranges set forth herein are exemplary only and are not intended to limit the scope or use of compositions and combinations containing them. The compositions can be administered hourly, daily, weekly, monthly, yearly, or once. Generally, dosage regimens are chosen to limit toxicity. It should be noted that the attending physician would know how to and when to terminate, interrupt, or adjust therapy to lower dosage due to toxicity, or bone marrow, liver, or kidney, or other tissue dysfunctions. Conversely, the attending physician would also know how to and when to adjust treatment to higher levels if the clinical response is not adequate (precluding toxic side effects).

The immunostimulatory bacteria are included in the composition in an amount sufficient to exert a therapeutically useful effect. For example, the amount is one that achieves a therapeutic effect in the treatment of a hyperproliferative disease or condition, such as cancer, or an amount that is effective as a vaccine, or to deliver mRNA. An exemplary dose can be about 1 x 10 ⁹ CFUs/m ² , but depends upon the route of administration. As shown in the table below, and noted above, higher doses can be administered. The data below, from an experiment in a mouse model, show that strains with the genome modifications as described herein have significantly improved tolerability, at least about 15-fold, compared to strain VNP20009, and, thus, can be dosed in higher amounts.

The immunostimulatory bacteria described and provided herein can be administered by any suitable route, including, but not limited to, intravenously (IV), and mucosally, such as by nasal/intranasal administration, and by inhalation into the lung. This is particularly advantageous in embodiments in which the bacteria are used as vaccines against pathogens, such as bacterial and viral pathogens, including, for example, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). The bacteria can be provided as a tablet, powder, or liquid, or other suitable formulation; they can be stored frozen, or refrigerated, or stored at room temperature, depending upon the formulation. Thus, the bacteria that provide mRNA can be administered via a variety of routes, and can be conveniently stored and formulated.

Pharmaceutically and therapeutically active compounds and derivatives thereof are typically formulated and administered in unit dosage forms or multiple dosage forms. Each unit dose contains a predetermined quantity of therapeutically active compound sufficient to produce the desired therapeutic effect, in association with the required pharmaceutical carrier, vehicle, or diluent. Unit dosage forms, include, but are not limited to, tablets, capsules, pills, powders, granules, parenteral suspensions, oral solutions or suspensions, and oil-in-water emulsions, containing suitable quantities of the compounds or pharmaceutically acceptable derivatives thereof. Unit dose forms can be contained in vials, ampoules and syringes, or individually packaged tablets or capsules. Unit dose forms can be administered in fractions or multiples thereof. A multiple dose form is a plurality of identical unit dosage forms packaged in a single container to be administered in segregated unit dose form. Examples of multiple dose forms include vials, bottles of tablets or capsules, or bottles of pints or gallons. Hence, multiple dose form is a multiple of unit doses that are not segregated in packaging. Generally, dosage forms or compositions containing active ingredient in the range of 0.005% to 100%, with the balance made up from non-toxic carrier, can be prepared. Pharmaceutical compositions can be formulated in dosage forms appropriate for each route of administration.

The unit-dose parenteral preparations are packaged in an ampoule, a vial, or a syringe with a needle. The volume of liquid solution or reconstituted powder preparation, containing the pharmaceutically active compound, is a function of the disease to be treated and the particular article of manufacture chosen for package. All preparations for parenteral administration must be sterile, as is known and practiced in the art.

As indicated, compositions provided herein can be formulated for any route known to those of skill in the art, including, but not limited to, subcutaneous, intramuscular, intravenous, intradermal, intralesional, intraperitoneal, epidural, vaginal, rectal, local, otic, or transdermal administration, or any route of administration. Formulations suited for such routes are known to one of skill in the art. Compositions also can be administered with other biologically active agents, either sequentially, intermittently, or in the same composition.

Pharmaceutical compositions can be administered by controlled release formulations and/or delivery devices (see, e.g., U.S. Patent Nos. 3,536,809; 3,598,123; 3,630,200; 3,845,770; 3,847,770; 3,916,899; 4,008,719; 4,687,660; 4,769,027; 5,059,595; 5,073,543; 5,120,548; 5,354,556; 5,591,767; 5,639,476; 5,674,533; and 5,733,566). Various delivery systems are known and can be used to administer selected compositions, are contemplated for use herein, and such particles can be easily made.

6. Packaging and Articles of Manufacture

Also provided are articles of manufacture containing packaging materials, any pharmaceutical composition provided herein, and a label that indicates that the compositions are to be used for treatment of diseases or conditions as described herein. For example, the label can indicate that the treatment is for a tumor or for cancer.

Combinations of immunostimulatory bacteria described herein and another therapeutic agent also can be packaged in an article of manufacture. In one example, the article of manufacture contains a pharmaceutical composition containing the immunostimulatory bacteria composition and no further agent or treatment. In other examples, the article of manufacture contains another further therapeutic agent, such as a different anti-cancer agent. In this example, the agents can be provided together or separately, for packaging as articles of manufacture.

The articles of manufacture provided herein contain packaging materials. Packaging materials for use in packaging pharmaceutical products are well-known to those of skill in the art. See, for example, U.S. Patent Nos. 5,323,907, 5,052,558, and 5,033,252, each of which is incorporated herein in its entirety. Examples of pharmaceutical packaging materials include, but are not limited to, blister packs, bottles, tubes, inhalers, pumps, bags, vials, containers, syringes, bottles, and any packaging material suitable for a selected formulation and intended mode of administration and treatment. Exemplary of articles of manufacture are containers, including single chamber and dual chamber containers. The containers include, but are not limited to, tubes, bottles, and syringes. The containers can further include a needle for intravenous administration.

The choice of package depends on the agents, and whether such compositions will be packaged together or separately. In general, the packaging is non-reactive with the compositions contained therein. In other examples, some of the components can be packaged as a mixture. In other examples, all components are packaged separately. Thus, for example, the components can be packaged as separate compositions that, upon mixing just prior to administration, can be directly administered together. Alternatively, the components can be packaged as separate compositions for administration separately.

Selected compositions including articles of manufacture thereof also can be provided as kits. Kits can include a pharmaceutical composition described herein, and an item for administration provided as an article of manufacture. The compositions can be contained in the item for administration or can be provided separately to be added later. The kit can, optionally, include instructions for application, including dosages, dosing regimens, and instructions for modes of administration. Kits also can include a pharmaceutical composition described herein and an item for diagnosis.

I. METHODS OF TREATMENT AND USES

The methods provided herein include methods of administering or using the immunostimulatory bacteria, for treating subjects having a disease or condition whose symptoms can be ameliorated or lessened by administration of such bacteria, such as cancer. In particular examples, the disease or condition is a tumor or a cancer. Bacteria provided herein also can be used as vaccines and other such applications, discussed in sections above. In those instances, not only can the bacteria be administered as described in this section, as described with reference to their use as anti-cancer agents, but also by other routes that mimic the natural routes of entry of various pathogens. For example, as discussed in the above sections, for vaccination, such as against viruses, such as respiratory viruses, they can be administered by inhalation into the nose or lungs. These bacteria elicit T-cell responses and immune responses in the tissues in which they are administered, such as the nose, esophagus, and lungs, to promote in situ immune responses, resulting in memory T-cells that are antigen-specific, as well as B-cells, to provide long-term responses.

Additionally, methods of combination therapies with one or more additional agents for treatment, such as an anti -cancer agent or an anti-hyaluronan agent, also are provided. The bacteria can be administered by any suitable route, including, but not limited to, parenteral, systemic, topical, and local, such as intra-tumoral, intravenous, rectal, oral, intramuscular, mucosal, and other routes. Because of the modifications of the bacteria described herein, problems associated with systemic administration are solved. Formulations suitable for each route of administration are provided. The skilled person can establish suitable regimens and doses, and select routes of administration.

1. Diagnostics for Patient Selection for Treatment and for

Monitoring Treatment a. Patient Selection

Biomarkers can be used to identify patients who are likely to respond to therapy with the immunostimulatory bacteria provided herein. For example, the Adenosine Signature and the Myeloid Signature can be assessed by NanoString gene expression panels, and T-cell infiltration of tumors can be assessed by the Immunoscore® test, which is an in vitro diagnostic test used for predicting the risk of relapse in early stage colon cancer patients, by measuring the host immune response at a tumor site. Patients whose tumors or body fluids indicate an immune responsiveness or an immune response are more likely to respond to the treatment with the immunostimulatory bacteria provided herein.

Other biomarkers include tumor-infiltrating lymphocytes (TILs), CD73, CD39, TNAP (tissue-nonspecific alkaline phosphatase), CD38, CD68, PD-L1, and FoxP3. For example, tumors that can be treated with the immunostimulatory bacteria provided herein are T-cell excluded, exhibit high levels of purines/ad enosine, and are unresponsive to PD-1/PD-L1 targeted therapies.

Gene expression profiles (GEPs), which can be determined using various NanoString gene expression panels, can be analyzed, for example, to identify the “adenosine signature” of tumors. High concentrations of adenosine are found in certain tumors, including colorectal carcinoma (CRC), non-small cell lung cancer (NSCLC), and pancreatic cancer, among others. Patients with tumors that exhibit high concentrations of purines/adenosine are likely to respond to therapy, since the immunostimulatory bacteria herein accumulate and replicate in purine/adenosine rich tumor microenvironments. Thus, the identification of tumors that express an “Adenosine Signature” can be used to predict patient response to therapy. Additionally, the immunostimulatory bacteria herein preferentially accumulate in and infect tumor-resident myeloid cells. Thus, the “Myeloid Signature” also can be used to predict patient response to therapy with the immunostimulatory bacteria. For example, it has been shown that the “Adenosine Signature” is nearly identical to the “Myeloid Signature” that is associated with poor response to atezolizumab (anti-PD- Ll) monotherapy in renal cell carcinoma (RCC) patients, which is indicative of the role of adenosine in tumor escape from anti-PD-Ll therapy (see, e.g., McDermott et al. (2018) Nature Medicine 24:749-757). Tumor-myeloid and tumor-adenosine NanoString signature panels are available and can be used for the selection of patients.

Macrophages limit T-cell infiltration into solid tumors and suppress their function, for example, in triple negative breast cancer (see, e.g., Keren et al. (2018) Cell 174:1373-1387). In certain cancers, such as CRC, macrophages dominate the intratumoral immune population and promote T-cell exclusion, and as a result, tumor- associated macrophages are associated with poor prognosis in CRC (see, e.g., Bindea et al. (2013) Immunity 39:782-795). Immunoscore®, a method to estimate the prognosis of cancer patients, based on the immune cells that infiltrate the cancer and surround it, can be used to measure T-cell exclusion or T-cell infiltration.

Immunoscore® incorporates the effects of the host immune response into cancer classification and improves prognostic accuracy. It measures the density of two T lymphocyte populations (CD3/CD8, CD3/CD45RO, or CD8/CD45RO) in the center and at the periphery of the tumor, and provides a score ranging from 0 (10), when low densities of both cell types are found in both regions, to Immunoscore 4 (14), when high densities are found in both regions. Low infiltration of T lymphocytes results in a low Immunoscore®, which correlates with high risk, while high infiltration of T lymphocytes results in a high Immunoscore®, which correlates with low risk. Immunoscore®, thus, can be evaluated as a prospective biomarker to identify patients that will respond to therapy with the immunostimulatory bacteria provided herein. For example, T-cell poor/uninflamed tumors can be treated, because the immunostimulatory bacteria provided herein induce T-cell infiltration in cold tumors. Such tumors represent a high, unmet need population that is refractory to checkpoint inhibition.

Extracellular adenosine is produced by the sequential activities of membrane associated ectoenzymes, CD39 (ecto-nucleoside triphosphate diphosphohydrolase 1, orNTPDasel) and CD73 (ecto-5’ -nucleotidase), which are expressed on tumor stromal cells, together producing adenosine by phosphohydrolysis of ATP or ADP that is produced from dead or dying cells. CD39 converts extracellular ATP (or ADP) to 5 ’-AMP, which is converted to adenosine by CD73. Expression of CD39 and CD73 on endothelial cells is increased under the hypoxic conditions of the tumor microenvironment, thereby increasing levels of adenosine. Thus, CD39 and CD73 can be used as biomarkers that indicate adenosine-rich tumors that can be targeted with the immunostimulatory bacteria provided herein.

CD38, also known as cyclic ADP ribose hydrolase, is a glycoprotein that is found on the surface of many immune cells, including CD4 ⁺ T-cells, CD8 ⁺ T-cells, B- lymphocytes, and natural killer cells. The loss of CD38, which is a marker of cell activation, is associated with impaired immune responses, and has been linked to leukemias, myelomas, and solid tumors. Additionally, increased expression of CD38 is an unfavorable diagnostic marker in chronic lymphocytic leukemia and is associated with increased disease progression. CD38 also is used as a target for daratumumab (Darzalex®), which has been approved for the treatment of multiple myeloma. CD68 is highly expressed by monocytes, circulating macrophages and by tissue macrophages (e.g., Kupffer cells, microglia). FoxP3 is involved in immune system responses, and acts as a regulator in the development and function of regulatory T-cells (or Tregs), which are immunosuppressive. In cancer, an excess of regulatory T-cell activity can prevent the immune system from destroying cancer cells. Thus, CD38, CD68 and FoxP3 also can be used as biomarkers for the selection of patients that are likely, to respond to therapy with the immunostimulatory bacteria herein. b. Diagnostics to Assess or Detect Activity of the Immunostimulatory Bacteria are Indicative of the Effectiveness of Treatment

Biomarkers can be used to monitor the immunostimulatory bacteria following treatment. Biomarkers occur in tumor samples and/or in body fluid samples, such as blood, plasma, urine, saliva, and other fluids. Validated, peripheral blood biomarkers are used to evaluate the immune status of patients prior to and during treatment, to determine changes in the immune status, which correlate with the effectiveness of treatment. A change to, or an increase in, anti-tumor immune response status indicates that treatment with the immunostimulatory bacteria is having an effect. Immunomodulatory activity of the immunostimulatory bacteria provided herein, for example, in dose escalation and expansion studies, can be assessed. Examination of biomarkers reveals prognostic and predictive factors relating to disease (e.g., a tumor) status and its treatment, which can aid in monitoring treatment. Evaluating the tumor microenvironment, for example, provides insights into the mechanism of tumor responses to immunotherapies. Serum biomarkers to detect immunomodulatory activity of the immunostimulatory bacteria include, but are not limited to, CXCL10 (IP- 10), CXCL9, interferon-p, interferon-y, proinflammatory serum cytokines (e.g., IL-6, TNF-a, MCP-1/CCL1), and IL-18 binding protein.

CXCL10 and CXCL9 are chemokines that are necessary for CD8 ⁺ T-cell activation and trafficking to tumors, for example, in response to immunotherapies. In a phase 3 trial of nivolumab, an anti-PD-1 immune checkpoint inhibitor, for the treatment of previously treated patients with metastatic renal cell carcinoma (mRCC), immune pharmacodynamic effects that were shared by the majority of patients, irrespective of the dose administered, were identified. Assessment of the IFN-y regulated serum chemokines CXCL9 and CXCL10 was performed using a multiplex panel based on Luminex technology (Myriad® Rules-Based Medicine (RBM)), and the results demonstrated that increased CXCL9 and CXCL10 serum levels, as well as increased transcription in the tumor, correlated with clinical response Median increases in chemokine levels after treatment with nivolumab from baseline were 101% for CXCL9 and 37% for CXCL10 in peripheral blood (see, e.g., Choueiri et al. (2016) Clin. Cancer Res. 22(22): 5461 -5471 ). Additionally, treatment of patients with advanced solid tumors or lymphomas with the MK-1454 STING agonist (Merck), resulted in a dose-dependent increase in serum CXCL10 after intratumoral dosing (see, e.g., Harrington et al. ESMO Annual Meeting (2018)).

Dose-dependent increases in serum levels of IFN- were observed following intratumoral dosing of the ADU-S100 STING agonist (Aduro) (see, e.g., Meric- Bemstam etal. ASCO Annual Meeting (2019)). Additionally, intravenous administration of VNP20009 induced a dose-dependent increase in the serum levels of the pro-inflammatory cytokines IL-6, TNF-a, IL-ip, and IL-12 (see, e.g., Toso et al. (2002) J. Clin. Oncol. 20(1): 142-152).

IL-18 participates in protective immune responses to intracellular bacteria, fungi and viruses, and has demonstrated anti-tumor activity in preclinical models of lung cancer, breast cancer, sarcoma, and melanoma. The biological activity of IL-18 is modulated in a negative feedback loop by IL-18 binding protein (IL-18BP), induced through IFN-y. Thus, serum levels of IL-18BP are predictive of clinical IFN- y activity. The intravenous administration of recombinant human IL-18 (rhIL-18) to patients with advanced cancers resulted in increased serum concentrations of IL- 18 binding protein in a dose-dependent manner, as well as increases in IFN-y, GM-CSF, and soluble Fas ligand (see, e.g., Robertson et al. (2006) Clin. Cancer Res. 12(14):4265-4273). Additionally, the levels of IL-18BP in urine and serum were observed to correlate with tumor status in patients with prostate cancer; significant differences in urinary IL-18BP levels were found between cases with and without prostate cancer, and increased serum IL-18BP levels correlated with increasing prostate cancer Gleason score, demonstrating that elevated IL-18BP secretion from prostate cancer cells can be indicative of an attempt by cancer to escape immune surveillance (see, e.g., Fujita etal. (2011) Ait. J. Cancer 129(2):424-432). Thus, IL- 18BP can be used as a biomarker for tumor immune responses. 2. Tumors

The immunostimulatory bacteria, combinations, uses and methods provided herein are applicable to treating all types of tumors, including cancers, particularly solid tumors, including lung cancer, bladder cancer, non-small cell lung cancer, gastric cancers, head and neck cancers, ovarian cancer, liver cancer, pancreatic cancer, kidney cancer, breast cancer, colorectal cancer, and prostate cancer. The methods also can be used for treating hematological cancers.

Tumors and cancers subject to treatment by the immunostimulatory bacteria, compositions, combinations, uses and methods provided herein include, but are not limited to, those that originate in the immune system, skeletal system, muscles and heart, breast, pancreas, gastrointestinal tract, central and peripheral nervous system, renal system, reproductive system, respiratory system, skin, connective tissue systems, including joints, fatty tissues, and the circulatory system, including blood vessel walls. Examples of tumors that can be treated with the immunostimulatory bacteria provided herein include carcinomas, gliomas, sarcomas (including liposarcoma), adenocarcinomas, adenosarcomas, and adenomas. Such tumors can occur in virtually all parts of the body, including, for example, the breast, heart, lung, small intestine, colon, spleen, kidney, bladder, head and neck, ovary, prostate, brain, pancreas, skin, bone, bone marrow, blood, thymus, uterus, testicles, cervix, or liver.

Tumors of the skeletal system include, for example, sarcomas and blastomas such as osteosarcoma, chondrosarcoma, and chondroblastoma. Muscle and heart tumors include tumors of both skeletal and smooth muscles, e.g., leiomyomas (benign tumors of smooth muscle), leiomyosarcomas, rhabdomyomas (benign tumors of skeletal muscle), rhabdomyosarcomas, and cardiac sarcomas. Tumors of the gastrointestinal tract include, e.g., tumors of the mouth, esophagus, stomach, small intestine, and colon, and colorectal tumors, as well as tumors of gastrointestinal secretory organs, such as the salivary glands, liver, pancreas, and the biliary tract. Tumors of the central nervous system (CNS) include tumors of the brain, retina, and spinal cord, and can also originate in associated connective tissue, bone, blood vessels, or nervous tissue. Treatment of tumors of the peripheral nervous system are also contemplated. Tumors of the peripheral nervous system include malignant peripheral nerve sheath tumors. Tumors of the renal system include those of the kidneys, e.g., renal cell carcinoma, as well as tumors of the ureters and bladder. Tumors of the reproductive system include tumors of the cervix, uterus, ovary, prostate, testes, and related secretory glands. Tumors of the immune system include both blood-based and solid tumors, including lymphomas, e.g., both Hodgkin’s and non-Hodgkin’s lymphomas. Tumors of the respiratory system include tumors of the nasal passages, bronchi, and lungs. Tumors of the breast include, e.g., both lobular and ductal carcinomas.

Other examples of tumors that can be treated by the immunostimulatory bacteria and methods provided herein include Kaposi’s sarcoma, CNS neoplasms, neuroblastomas, capillary hemangioblastomas, meningiomas and cerebral metastases, melanoma, gastrointestinal and renal carcinomas and sarcomas, rhabdomyosarcoma, glioblastoma (such as glioblastoma multiforme), and leiomyosarcoma. Examples of other cancers that can be treated as provided herein include, but are not limited to, lymphoma, blastoma, neuroendocrine tumors, mesothelioma, schwannoma, meningioma, melanoma, and leukemia or lymphoid malignancies. Examples of such cancers include hematologic malignancies, such as Hodgkin’s lymphoma, nonHodgkin’s lymphomas (Burkitt’s lymphoma, small lymphocytic lymphoma, chronic lymphocytic leukemia, mycosis fungoides, mantle cell lymphoma, follicular lymphoma, diffuse large B-cell lymphoma, marginal zone lymphoma, hairy cell leukemia, and lymphoplasmacytic leukemia), tumors of lymphocyte precursor cells, including B-cell acute lymphoblastic leukemia/lymphoma, and T-cell acute lymphoblastic leukemia/lymphoma, thymoma, tumors of the mature T and NK cells, including peripheral T-cell leukemias, adult T-cell leukemia/T-cell lymphomas and large granular lymphocytic leukemia, Langerhans cell histiocytosis, myeloid neoplasias such as acute myelogenous leukemias, including acute myeloid leukemia (AML) with maturation, AML without differentiation, acute promyelocytic leukemia, acute myelomonocytic leukemia, and acute monocytic leukemias, myelodysplastic syndromes, and chronic myeloproliferative disorders, including chronic myelogenous leukemia; tumors of the central nervous system such as glioma, glioblastoma, neuroblastoma, astrocytoma, medulloblastoma, ependymoma, and retinoblastoma; solid tumors of the head and neck (e.g., nasopharyngeal cancer, salivary gland carcinoma, and esophageal cancer), lung (e.g., small-cell lung cancer, non-small cell lung cancer, adenocarcinoma of the lung, and squamous carcinoma of the lung), digestive system (e.g., gastric or stomach cancer, including gastrointestinal cancer, cancer of the bile duct or biliary tract, colon cancer, rectal cancer, colorectal cancer, and anal carcinoma), reproductive system (e.g., testicular, penile, prostate, uterine, vaginal, vulval, cervical, ovarian, and endometrial cancers), skin (e.g., melanoma, basal cell carcinoma, squamous cell cancer, actinic keratosis, and cutaneous melanoma), liver (e.g., liver cancer, hepatic carcinoma, hepatocellular cancer, and hepatoma), bone (e.g., osteoclastoma, and osteolytic bone cancers), additional tissues and organs (e.g., pancreatic cancer, bladder cancer, kidney or renal cancer, thyroid cancer, breast cancer, cancer of the peritoneum, and Kaposi’s sarcoma), tumors of the vascular system (e.g., angiosarcoma and hemangiopericytoma), Wilms’ tumor, retinoblastoma, osteosarcoma, and Ewing’s sarcoma.

3. Administration

In practicing the uses and methods herein, immunostimulatory bacteria provided herein can be administered to a subject, including a subject having a tumor or having neoplastic cells, or a subject to be immunized. One or more steps can be performed prior to, simultaneously with, or after administration of the immunostimulatory bacteria to the subject, including, but not limited to, diagnosing the subject with a condition appropriate for administering immunostimulatory bacteria, determining the immunocompetence of the subject, immunizing the subject, treating the subject with a chemotherapeutic agent, treating the subject with radiation, or surgically treating the subject.

For embodiments that include administering immunostimulatory bacteria to a tumor-bearing subject for therapeutic purposes, the subject typically has previously been diagnosed with a neoplastic condition. Diagnostic methods also can include determining the type of neoplastic condition, determining the stage of the neoplastic condition, determining the size of one or more tumors in the subject, determining the presence or absence of metastatic or neoplastic cells in the lymph nodes of the subject, or determining the presence of metastases in the subject.

Some embodiments of the therapeutic methods for administering immunostimulatory bacteria to a subject can include a step of determining the size of the primary tumor or the stage of the neoplastic disease, and, if the size of the primary tumor is equal to or above a threshold volume, or if the stage of the neoplastic disease is at or above a threshold stage, an immunostimulatory bacterium is administered to the subject. In a similar embodiment, if the size of the primary tumor is below a threshold volume, or if the stage of the neoplastic disease is at or below a threshold stage, the immunostimulatory bacterium is not yet administered to the subject; such methods can include monitoring the subject until the tumor size or neoplastic disease stage reaches a threshold amount, and then administering the immunostimulatory bacterium to the subject. Threshold sizes can vary according to several factors, including rate of growth of the tumor, ability of the immunostimulatory bacterium to infect a tumor, and immunocompetence of the subject. Generally, the threshold size will be a size sufficient for an immunostimulatory bacterium to accumulate and replicate in or near the tumor, without being completely removed by the host’s immune system, and will typically also be a size sufficient to sustain a bacterial infection for a time long enough for the host to mount an immune response against the tumor cells, typically about one week or more, about ten days or more, or about two weeks or more. Exemplary threshold stages are any stage beyond the lowest stage (e.g., Stage I or equivalent), or any stage where the primary tumor is larger than a threshold size, or any stage where metastatic cells are detected.

Any mode of administration of a microorganism to a subject can be used, provided the mode of administration permits the immunostimulatory bacteria to enter a tumor or metastasis. Modes of administration can include, but are not limited to, intravenous, intraperitoneal, subcutaneous, intramuscular, topical, intratumoral, multipuncture, inhalation, intranasal, oral, intracavity (e.g., administering to the bladder via a catheter, or administering to the gut by suppository or enema), aural, rectal, and ocular administration.

One skilled in the art can select any mode of administration compatible with the subject and the bacteria, and that also is likely to result in the bacteria reaching tumors and/or metastases. The route of administration can be selected by one skilled in the art according to any of a variety of factors, including the nature of the disease, the kind of tumor, and the particular bacteria contained in the pharmaceutical composition. Administration to the target site can be performed, for example, by ballistic delivery, or as a colloidal dispersion system, or systemic administration can be performed by injection into an artery.

The dosage regimen can be any of a variety of methods and amounts, and can be determined by one skilled in the art according to known clinical factors. A single dose can be therapeutically effective for treating a disease or disorder in which immune stimulation effects treatment. Exemplary of such stimulation is an immune response, that includes, but is not limited to, one or both of a specific immune response and non-specific immune response, both specific and non-specific responses, innate response, primary immune response, adaptive immunity, secondary immune response, memory immune response, immune cell activation, immune cell proliferation, immune cell differentiation, and cytokine expression.

As is known in the medical arts, dosages for a subject can depend on many factors, including the subject’s species, size, body surface area, age, sex, immunocompetence, and general health, the particular bacteria to be administered, the duration and route of administration, the kind and stage of the disease, for example, the tumor size, and other compounds, such as drugs, being administered concurrently. In addition to the above factors, such levels can be affected by the infectivity of the bacteria and the nature of the bacteria, as can be determined by one skilled in the art. In the present methods, appropriate minimum dosage levels of bacteria can be levels sufficient for the bacteria to survive, grow and replicate in a tumor or metastasis. Exemplary minimum levels for administering a bacterium to a 65 kg human can include at least about 5 x 10 ⁶ colony forming units (CFUs), at least about 1 x 10 ⁷ CFUs, at least about 5 x 10 ⁷ CFUs, at least about 1 x 10 ⁸ CFUs, or at least about 1 x 10 ⁹ CFUs. In the present methods, appropriate maximum dosage levels of bacteria can be levels that are not toxic to the host, levels that do not cause splenomegaly of 3x or more, or levels that do not result in colonies or plaques in normal tissues or organs after about 1 day, or after about 3 days, or after about 7 days. Exemplary maximum levels for administering a bacterium to a 65 kg human can include no more than about 5 x 10 ¹¹ CFUs, no more than about 1 x 10 ¹¹ CFUs, no more than about 5 x IO ¹⁰ CFUs, no more than about 1 x IO ¹⁰ CFUs, or no more than about 1 x 10 ⁹ CFUs.

The methods and uses provided herein can include a single administration of immunostimulatory bacteria to a subject, or multiple administrations of immunostimulatory bacteria to a subject, or others of a variety of regimens, including combination therapies with other anti-tumor therapeutics and/or treatments. These include, for example, cellular therapies, such as administration of modified immune cells; CAR-T therapy; CRISPR therapy; checkpoint inhibitors, such as antibodies (e.g, anti-PD-1 antibodies, anti-PD-Ll antibodies, and anti-CTLA-4 antibodies, and other such immunotherapies); chemotherapeutic compounds, such as nucleoside analogs; surgery; and radiotherapy. Other cancer therapies also include anti-VEGF, anti-VEGFR, anti-VEGFR2, anti-TGF-P, or anti-IL-6 antibodies, or fragments thereof, cancer vaccines, and oncolytic viruses.

In some embodiments, a single administration is sufficient to establish immunostimulatory bacteria in a tumor, where the bacteria can colonize, and can cause or enhance an anti-tumor response in the subject. In other embodiments, the immunostimulatory bacteria provided for use in the methods herein can be administered on different occasions, separated in time, typically by at least one day Separate administrations can increase the likelihood of delivering a bacterium to a tumor or metastasis, where a previous administration may have been ineffective in delivering the bacterium to a tumor or metastasis. In embodiments, separate administrations can increase the locations on a tumor or metastasis where bacterial colonization/proliferation can occur, or can otherwise increase the titer of bacteria accumulated in the tumor, which can increase eliciting or enhancing a host’s antitumor immune response.

When separate administrations are performed, each administration can be a dosage amount that is the same or different relative to other administration dosage amounts. In one embodiment, all administration dosage amounts are the same. In other embodiments, a first dosage amount can be a larger dosage amount than one or more subsequent dosage amounts, for example, at least lOx larger, at least lOOx larger, or at least lOOOx larger, than subsequent dosage amounts. In one example of a method of separate administrations, in which the first dosage amount is greater than one or more subsequent dosage amounts, all subsequent dosage amounts can be the same, smaller amount, relative to the first administration.

Separate administrations can include any number of two or more administrations, including two, three, four, five, or six administrations. One skilled in the art readily can determine the number of administrations to perform, or the desirability of performing one or more additional administrations, according to methods known in the art for monitoring therapeutic methods, and other monitoring methods provided herein. Accordingly, the methods provided herein include methods of providing to the subject one or more administrations of immunostimulatory bacteria, where the number of administrations can be determined by monitoring the subject, and, based on the results of the monitoring, determining whether or not to provide one or more additional administrations. Deciding whether or not to provide one or more additional administrations can be based on a variety of monitoring results, including, but not limited to, indication of tumor growth or inhibition of tumor growth, appearance of new metastases or inhibition of metastasis, the subject’s antibacterial antibody titer, the subject’s anti -tumor antibody titer, the overall health of the subject, and the weight of the subject. The time period between administrations can be any of a variety of time periods. The time period between administrations can be a function of any of a variety of factors, including monitoring steps, as described in relation to the number of administrations, the time period for a subject to mount an immune response, the time period for a subject to clear bacteria from normal tissue, or the time period for bacterial colonization/proliferation in the tumor or metastasis. In one example, the time period can be a function of the time period for a subject to mount an immune response; for example, the time period can be more than the time period for a subject to mount an immune response, such as more than about one week, more than about ten days, more than about two weeks, or more than about a month. In another example, the time period can be less than the time period for a subject to mount an immune response, such as less than about one week, less than about ten days, less than about two weeks, or less than about a month. In another example, the time period can be a function of the time period for bacterial colonization/proliferation in the tumor or metastasis; for example, the time period can be more than the amount of time for a detectable signal to arise in a tumor or metastasis after administration of a microorganism expressing a detectable marker, such as about 3 days, about 5 days, about a week, about ten days, about two weeks, or about a month.

The methods used herein also can be performed by administering compositions, such as suspensions and other formulations, containing the immunostimulatory bacteria provided herein. Such compositions contain the bacteria and a pharmaceutically acceptable excipient or vehicle, as provided herein or known to those of skill in the art.

As discussed above, the uses and methods provided herein also can include administering one or more therapeutic compounds, such as anti-tumor compounds or other cancer therapeutics, to a subject, in addition to administering the immunostimulatory bacteria to the subject. The therapeutic compounds can act independently, or in conjunction with the immunostimulatory bacteria, for tumor therapeutic effects. Therapeutic compounds that can act independently include any of a variety of known chemotherapeutic compounds that can inhibit tumor growth, inhibit metastasis growth and/or formation, decrease the size of a tumor or metastasis, or eliminate a tumor or metastasis, without reducing the ability of the immunostimulatory bacteria to accumulate in a tumor, replicate in the tumor, and cause or enhance an anti-tumor immune response in the subject. Examples of such chemotherapeutic agents include, but are not limited to, alkylating agents, such as thiotepa and cyclophosphamide; alkyl sulfonates, such as busulfan, improsulfan, and piposulfan; androgens, such as calusterone, dromostanolone propionate, epitiostanol, mepitiostane, and testolactone; anti-adrenals, such as aminoglutethimide, mitotane, and trilostane; anti-androgens, such as flutamide, nilutamide, bicalutamide, leuprolide, and goserelin; antibiotics, such as aclacinomycin, actinomycin, anthramycin, azaserine, bleomycin, cactinomycin, calicheamicin, carubicin, carminomycin, carzinophilin, chromomycin, dactinomycin, daunorubicin, detorubicin, 6-diazo-5-oxo-L-norleucine, doxorubicin, epirubicin, esorubicin, idarubicin, marcellomycin, mitomycins, mycophenolic acid, nogalamycin, olivomycins, peplomycin, porfiromycin, puromycin, quelamycin, rodorubicin, streptonigrin, streptozocin, tubercidin, ubenimex, zinostatin, and zorubicin; antiestrogens, including, for example, tamoxifen, raloxifene, aromatase inhibiting 4(5)- imidazoles, 4-hydroxytamoxifen, trioxifene, keoxifene, LY 117018, onapristone, and toremifene (Fareston®); anti-metabolites, such as methotrexate and 5 -fluorouracil (5- FU); folic acid analogues, such as denopterin, methotrexate, pteropterin, and trimetrexate; aziridines, such as benzodepa, carboquone, meturedepa, and uredepa; ethylenimines and methylmelamines, including altretamine, triethylenemelamine, triethylenephosphoramide, triethylenethiophosphoramide, and trimethylol melamine; folic acid replenishers, such as folinic acid; nitrogen mustards, such as chlorambucil, chlornaphazine, chlorophosphamide, estramustine, ifosfamide, mechlorethamine, mechlorethamine oxide hydrochloride, melphalan, novembichin, phenesterine, prednimustine, trofosfamide, and uracil mustard; nitrosoureas, such as carmustine, chlorozotocin, fotemustine, lomustine, nimustine, and ranimustine; platinum analogs, such as cisplatin and carboplatin; vinblastine; platinum; proteins, such as arginine deiminase and asparaginase; purine analogs, such as fludarabine, 6-mercaptopurine, thiamiprine, and thioguanine; pyrimidine analogs, such as ancitabine, azacitidine, 6- azauridine, carmofur, cytarabine, dideoxyuridine, doxifluridine, enocitabine, floxuridine, and 5-FU; taxanes, such as paclitaxel and docetaxel, and albuminated forms thereof (i.e., «czZ>-paclitaxel and //c/A-docetaxel); topoisomerase inhibitors, such as RFS-2000; thymidylate synthase inhibitors, such as Tomudex®; and additional chemotherapeutics, including aceglatone; aldophosphamide glycoside; aminolevulinic acid; amsacrine; bestrabucil; bisantrene; edatrexate; defosfamide; demecolcine; diaziquone; difluoromethylornithine (DFMO); eflornithine; elliptinium acetate; etoglucid; gallium nitrate; hydroxyurea; lentinan; lomdamine; mitoguazone; mitoxantrone; mopidamol; nitracrine; pentostatin; phenamet; pirarubicin; podophyllinic acid; 2-ethylhydrazide; procarbazine, PSK®; razoxane; sizofiran; spirogermanium; tenuazonic acid; triaziquone; 2,2’,2”-trichlorotriethylamine; urethan; vindesine; dacarbazine; mannomustine; mitobronitol; mitolactol; pipobroman; gacytosine; arabinoside (“Ara-C”); cyclophosphamide; thiotepa; chlorambucil; gemcitabine; 6-thioguanine; mercaptopurine; methotrexate; etoposide (VP- 16); ifosfamide; mitomycin C; mitoxantrone; vincristine; vinorelbine; Navelbine; Novantrone; teniposide; daunomycin; aminopterin; Xeloda®; ibandronate; CPT-11; retinoic acid; esperamycins; capecitabine; and topoisomerase inhibitors, such as irinotecan. Pharmaceutically acceptable salts, acids, or derivatives of any of the above also can be used.

Therapeutic compounds that act in conjunction with the immunostimulatory bacteria include, for example, compounds that increase the immune response eliciting properties of the bacteria, e.g., by increasing expression of encoded therapeutic products, such as cytokines, chemokines, co-stimulatory molecules, proteins that constitutively induce type I ZFNs, RNAi molecules that inhibit, suppress, or disrupt expression of checkpoint gene(s), a checkpoint inhibitor antibody and antibodies or fragments thereof against other targets, or compounds that can further augment bacterial colonization/proliferation. For example, a gene expression-altering compound can induce or increase transcription of a gene in a bacterium, such as an exogenous gene encoded on the plasmid, thereby provoking an immune response. Any of a wide variety of compounds that can alter gene expression are known in the art, including IPTG and RU486. Exemplary genes whose expression can be up- regulated include those encoding proteins and RNA molecules, including toxins, enzymes that can convert a prodrug to an anti-tumor drug, cytokines, transcription regulating proteins, shRNA, siRNA, and ribozymes. In other embodiments, therapeutic compounds that can act in conjunction with the immunostimulatory bacteria to increase the colonization/proliferation or immune response eliciting properties of the bacteria, are compounds that can interact with a bacterially-encoded gene product, and such interaction can result in an increased killing of tumor cells or an increased anti -tumor immune response in the subject. A therapeutic compound that can interact with a bacterially-encoded gene product can include, for example, a prodrug or other compound that has little or no toxicity, or other biological activity in its subject-administered form, but after interaction with a bactenally-encoded gene product, the compound can develop a property that results in tumor cell death, including but not limited to, cytotoxicity, the ability to induce apoptosis, or the ability to trigger an immune response. A variety of prodrug-like substances are known in the art, including ganciclovir, 5 -fluorouracil, 6-methylpurine deoxyriboside, cephalosporin-doxorubicin, 4-[(2-chloroethyl)(2-mesuloxyethyl)amino]benzoyl-L- glutamic acid, acetaminophen, indole-3-acetic acid, CB1954, 7-ethyl-l 0-[4-(l - piperidino)-l-piperidino]carbonyloxycamptothecin, bis-(2-chloroethyl)amino-4- hydroxyphenylaminomethanone 28, l-chloromethyl-5-hydroxy-l,2-dihyro-3H- benz[e]indole, epirubicin-glucuronide, 5’ -deoxy 5-fluorouri dine, cytosine arabinoside, and linamarin.

4. Monitoring

The methods provided herein can further include one or more steps of monitoring the subject, monitoring the tumor, and/or monitoring the immunostimulatory bacteria administered to the subject. Any of a variety of monitoring steps can be included in the methods provided herein, including, but not limited to, monitoring tumor size, monitoring the presence and/or size of metastases, monitoring the subject’s lymph nodes, monitoring the subject’s weight or other health indicators, including blood or urine markers, monitoring anti-bacterial antibody titer, monitoring bacterial expression of a detectable gene product, and directly monitoring bacterial titer in a tumor, tissue, or organ of a subject.

The purpose of the monitoring can be simply for assessing the health state of the subject, or the progress of therapeutic treatment of the subject, or can be for determining whether or not further administration of the same or a different immunostimulatory bacterium is warranted, or for determining when or whether or not to administer a compound to the subject where the compound can act to increase the efficacy of the therapeutic method, or the compound can act to decrease the pathogenicity of the bacteria administered to the subject.

In some embodiments, the methods provided herein can include monitoring one or more bacterially-expressed genes. Bacteria, such as those provided herein or otherwise known in the art, can express one or more detectable gene products, including but not limited to, detectable proteins.

As provided herein, measurement of a detectable gene product expressed in a bacterium can provide an accurate determination of the level of bacteria present in the subject. As further provided herein, measurement of the location of the detectable gene product, for example, by imaging methods, including tomographic methods, can determine the localization of the bacteria in the subject. Accordingly, the methods provided herein that include monitoring a detectable bacterial gene product can be used to determine the presence or absence of the bacteria in one or more organs or tissues of a subject, and/or the presence or absence of the bacteria in a tumor or metastases of a subject. Further, the methods provided herein that include monitoring a detectable bacterial gene product can be used to determine the titer of bacteria present in one or more organs, tissues, tumors, or metastases. Methods that include monitoring the localization and/or titer of bacteria in a subject can be used for determining the pathogenicity of bacteria, since bacterial infection, and particularly the level of infection, of normal tissues and organs can indicate the pathogenicity of the bacteria. The methods that include monitoring the localization and/or titer of the immunostimulatory bacteria in a subject can be performed at multiple time points and, accordingly, can determine the rate of bacterial replication in a subject, including the rate of bacterial replication in one or more organs or tissues of a subject; accordingly, methods that include monitoring a bacterial gene product can be used for determining the replication competence of the bacteria. The methods provided herein also can be used to quantitate the amount of immunostimulatory bacteria present in a variety of organs or tissues, and tumors or metastases, and can thereby indicate the degree of preferential accumulation of the bacteria in a subject; accordingly, the bacterial gene product monitoring can be used in methods of determining the ability of the bacteria to accumulate in tumors or metastases, in preference to normal tissues or organs. Since the immunostimulatory bacteria used in the methods provided herein can accumulate in an entire tumor or can accumulate at multiple sites in a tumor, and can also accumulate in metastases, the methods provided herein for monitoring a bacterial gene product can be used to determine the size of a tumor, or the number of metastases present in a subject. Monitoring such presence of a bacterial gene product in a tumor or metastasis over a range of time can be used to assess changes in the tumor or metastasis, including growth or shrinking of a tumor, or development of new metastases, or disappearance of metastases, and also can be used to determine the rate of growth or shrinking of a tumor, or the rate of development of new metastases or disappearance of metastases, or the change in the rate of growth or shrinking of a tumor, or the change in the rate of development of new metastases or disappearance of metastases. Accordingly, monitoring a bacterial gene product can be used for monitoring a neoplastic disease in a subject, or for determining the efficacy of treatment of a neoplastic disease, by determining the rate of growth or shrinking of a tumor, or the development of new metastases or disappearance of metastases, or the change in the rate of growth or shrinking of a tumor, or the development of new metastases or disappearance of metastases.

Any of a variety of detectable proteins can be detected by monitoring, exemplary of which are any of a variety of fluorescent proteins (e.g., green fluorescent proteins), any of a variety of luciferases, transferrin, or other iron-binding proteins; or receptors, binding proteins, and antibodies, where a compound that specifically binds the receptor, binding protein or antibody can be a detectable agent, or can be labeled with a detectable substance (e.g., a radionuclide or imaging agent).

Tumor and/or metastasis size can be monitored by any of a variety of methods known in the art, including external assessment methods, or tomographic or magnetic imaging methods. In addition to the methods known in the art, methods provided herein, for example, monitoring bacterial gene expression, can be used for monitoring tumor and/or metastasis size.

Monitoring size over several time points can provide information regarding the increase or decrease in the size of a tumor or metastasis, and can also provide information regarding the presence of additional tumors and/or metastases in the subject. Monitoring tumor size over several time points can provide information regarding the development of a neoplastic disease in a subject, including the efficacy of treatment of a neoplastic disease in a subject.

The methods provided herein also can include monitoring the antibody titer in a subject, including antibodies produced in response to administration of the immunostimulatory bacteria to a subject. The bacteria administered in the methods provided herein can elicit an immune response to endogenous bacterial antigens. The bacteria administered in the methods provided herein also can elicit an immune response to exogenous genes expressed by the bacteria. The bacteria administered in the methods provided herein also can elicit an immune response to tumor antigens. Monitoring antibody titer against bacterial antigens, bacterially-expressed exogenous gene products, or tumor antigens can be used to monitor the toxicity of the bacteria, to monitor the efficacy of treatment methods, or to monitor the level of gene product(s) or antibodies for production and/or harvesting. Monitoring antibody titer can be used to monitor the toxicity of the bacteria. Antibody titer against a bacteria can vary over the time period after administration of the bacteria to the subject, where at some particular time points, a low anti-(bacterial antigen) antibody titer can indicate a lower toxicity, while at other time points, a high anti-(bacterial antigen) antibody titer can indicate a higher toxicity. The bacteria used in the methods provided herein can be immunogenic, and can, therefore, elicit an immune response soon after administering the bacteria to the subject. Generally, immunostimulatory bacteria against which the immune system of a subject can mount a strong immune response can be bacteria that have low toxicity when the subject’s immune system can remove the bacteria from all normal organs or tissues. Thus, in some embodiments, a high antibody titer against bacterial antigens soon after administering the bacteria to a subject can indicate low toxicity of the bacteria.

In other embodiments, monitoring antibody titer can be used to monitor the efficacy of treatment methods. In the methods provided herein, antibody titer, such as anti-(tumor antigen) antibody titer, can indicate the efficacy of a therapeutic method, such as a therapeutic method to treat neoplastic disease. Therapeutic methods provided herein can include causing or enhancing an immune response against a tumor and/or metastasis. Thus, by monitoring the anti-(tumor antigen) antibody titer, it is possible to monitor the efficacy of a therapeutic method in causing or enhancing an immune response against a tumor and/or metastasis.

In other embodiments, monitoring antibody titer can be used for monitoring the level of gene product(s) or antibodies for production and/or harvesting. As provided herein, methods can be used for producing proteins, RNA molecules or other compounds, by expressing an exogenous gene in a microorganism that has accumulated in a tumor, in the tumor microenvironment, and/or in tumor-resident immune cells. Monitoring antibody titer against the protein, RNA molecule, or other compound can indicate the level of production of the protein, RNA molecule, or other compound by the tumor-accumulated microorganism, and also, can directly indicate the level of antibodies specific for such a protein, RNA molecule, or other compound.

The methods provided herein also can include methods of monitoring the health of a subject. Some of the methods provided herein are therapeutic methods, including neoplastic disease therapeutic methods. Monitoring the health of a subject can be used to determine the efficacy of the therapeutic method, as is known in the art. The methods provided herein also can include a step of administering to a subject an immunostimulatory bacterium, as provided herein. Monitoring the health of a subject can be used to determine the pathogenicity of an immunostimulatory bacterium administered to a subject. Any of a variety of health diagnostic methods for monitoring disease, such as neoplastic disease, infectious disease, or immune-related disease, can be monitored, as is known in the art. For example, the weight, blood pressure, pulse, breathing, color, temperature, or other observable state of a subject can indicate the health of a subject. In addition, the presence, or absence, or level of one or more components in a sample from a subject can indicate the health of a subject. Typical samples can include blood and urine samples, where the presence, or absence, or level of one or more components can be determined by performing, for example, a blood panel or a urine panel diagnostic test. Exemplary components indicative of a subject’s health include, but are not limited to, white blood cell count, hematocrit, and c-reactive protein concentration.

The methods provided herein can include monitoring a therapy, where therapeutic decisions can be based on the results of the monitoring. Therapeutic methods provided herein can include administering to a subject immunostimulatory bacteria, where the bacteria can preferentially accumulate in a tumor, the tumor microenvironment, or in tumor-resident immune cells, and/or in metastases, and where the bacteria can cause or enhance an anti-tumor immune response. Such therapeutic methods can include a variety of steps, including multiple administrations of a particular immunostimulatory bacterium, administration of a second immunostimulatory bacterium, or administration of a therapeutic compound. Determination of the amount, timing, or type of immunostimulatory bacteria or compound to administer to the subject can be based on one or more results from monitoring the subject. For example, the antibody titer in a subject can be used to determine whether or not it is desirable to administer an immunostimulatory bacterium and, optionally, a compound, the quantity of bacteria and/or compound to administer, and the type of bacteria and/or compound to administer, where, for example, a low antibody titer can indicate the desirability of administering an additional immunostimulatory bacterium, a different immunostimulatory bacterium, and/or a therapeutic compound, such as a compound that induces bacterial gene expression, or a therapeutic compound that is effective independent of the immunostimulatory bacteria. In another example, the overall health state of a subject can be used to determine whether or not it is desirable to administer an immunostimulatory bacterium and, optionally, a compound, the quantity of bacterium and/or compound to administer, and the type of bacterium and/or compound to administer, where, for example, determining that the subject is healthy can indicate the desirability of administering additional bacteria, different bacteria, or a therapeutic compound, such as a compound that induces bacterial gene/genetic payload/therapeutic product expression. In another example, monitoring a detectable bacterially-expressed gene product can be used to determine whether it is desirable to administer an immunostimulatory bacterium and, optionally, a compound, the quantity of bacterium and/or compound to administer, and the type of bacterium and/or compound to administer, where, for example, determining that the subject is healthy can indicate the desirability of administering additional bacteria, different bacteria, or a therapeutic compound, such as a compound that induces bacterial gene/genetic payload/therapeutic product expression. Such monitoring methods can be used to determine whether or not the therapeutic method is effective, whether or not the therapeutic method is pathogenic to the subject, whether or not the bacteria have accumulated in a tumor or metastasis, and whether or not the bacteria have accumulated in normal tissues or organs. Based on such determinations, the desirability and form of further therapeutic methods can be derived.

In another example, monitoring can determine whether or not immunostimulatory bacteria have accumulated in a tumor or metastasis of a subject. Upon such a determination, a decision can be made to further administer additional bacteria, a different immunostimulatory bacterium, and, optionally, a compound to the subject.

J. EXAMPLES

The following examples are included for illustrative purposes only and are not intended to limit the scope of the invention.

EXAMPLE 1 Auxotrophic Strains of S. typhimurium The Salmonella Strain YS1646 is Auxotrophic for Adenosine

Strains provided herein are engineered to be auxotrophic for adenosine. As a result, they are attenuated in vivo because they are unable to replicate in the low adenosine concentrations of normal tissue, and colonization occurs primarily in the solid tumor microenvironment (TME), where adenosine levels are high. The Salmonella strain YS1646 is a derivative of the wild-type strain ATCC 14028, and was engineered to be auxotrophic for purines due to disruption of the purl gene (synonymous with purM) (see, e.g., Low et al. (2004) Methods Mol. Med. 90:47-60). Subsequent analysis of the entire genome of YS1646 demonstrated that the purl gene was not in fact deleted, but was instead disrupted by a chromosomal inversion (see, e.g., Broadway et al. (2014) J. Biotechnol. 192: 177-178), and that the entire gene is still contained within two parts of the YS1646 chromosome that is flanked by insertion sequences, one of which has an active transposase. The presence of the complete genetic sequence of the purl gene, disrupted by means of a chromosomal reengagement, leaves open the possibility of reversion to a wild-type gene. While it has previously been demonstrated that the purine auxotrophy of YS 1646 was stable after >140 serial passages in vitro, it was not clear what the reversion rate is (see, e.g., Clairmont et al. (2000) J. Infect. Dis. 181: 1996-2002).

It is shown herein that, when provided with adenosine, strain YS1646 is able to replicate in minimal medium, whereas the wild-type parental strain, ATCC 14028 can grow in minimal media that is not supplemented with adenosine. Strain YS1646 was grown overnight in lysogeny broth (LB) medium, washed with M9 minimal medium, and diluted into M9 minimal medium containing no adenosine, or increasing concentrations of adenosine. Growth was measured using a SpectraMax® M3 Spectrophotometer (Molecular Devices) at 37 °C, reading the ODeoo every 15 minutes.

The results showed that, unlike a wild-type strain (ATCC 14028), which was able to grow in all concentrations of adenosine, strain YS1646 only was able to replicate when adenosine was provided at concentrations ranging from 11 to 300 micromolar, and was completely unable to replicate in M9 alone, or in M9 supplemented with 130 nanomolar adenosine. These data demonstrate that purl mutants are able to replicate at concentrations of adenosine that are found in the tumor microenvironment, but not at concentrations found in normal tissues. Engineered adenosine auxotrophic strains exemplified herein include strains in which all or portions of the purl open reading frame are deleted from the chromosome to prevent reversion to wild-type. Such gene deletions can be achieved by any method known to one of skill in the art, including the lambda red system, as described below.

The Salmonella Strain YS1646 is Auxotrophic for ATP In addition to the purine and adenosine auxotrophy, it was determined whether the /wAdcletcd strain also can scavenge ATP. ATP accumulates to high levels in the tumor microenvironment, due to leakage from dying tumor cells. It is shown herein that, when provided with ATP, strain YS1646 is able to replicate in minimal media, but is unable to grow when not supplemented with ATP. To demonstrate this, strain YS1646 was grown overnight in LB medium, washed with M9 minimal medium, and diluted into M9 minimal medium containing no ATP, or increasing concentrations of ATP (Fisher). Growth was measured using a SpectraMax® M3 Spectrophotometer (Molecular Devices) at 37 °C, reading the ODeoo every 15 minutes. The results demonstrated that strain YS1646 is able to replicate when ATP is provided at concentrations of 0.012 mM, but not in M9 alone.

EXAMPLE 2

Defects in Intracellular Replication are Attributed to the msbB Mutation

The YS1646 strain contains mutations in purl, which limits replication to sites containing high concentrations of purines, adenosine, or ATP, and mutations in msbB, which alters the lipopolysaccharide (LPS) surface coat in order to reduce TLR4- mediated pro-inflammatory signaling. It also has been established that, unlike wildtype Salmonella, strain YS1646 is unable to replicate in macrophages. Experiments were performed to determine which of these genetic mutations is responsible for conferring that phenotype within the wild-type strain, ATCC 14028.

In this assay, mouse RAW macrophage cells (InvivoGen, San Diego, Ca.) were infected with wild-type Salmonella strains containing deletions in purl, msbB, or both, at a multiplicity of infection (MOI) of approximately 5 bacteria per cell for 30 minutes, then the cells were washed with PBS, and medium containing gentamicin was added to kill extracellular bacteria. Intracellular bacteria are not killed by gentamicin, as it cannot cross the cell membrane. At various time points after infection, cell monolayers were lysed by osmotic shock with water, and the cell lysates were diluted and plated on LB agar to enumerate surviving colony forming units (CFUs).

As shown in the table below, wild-type Salmonella strains containing only the purl' mutation still were able to replicate. This explains why there is only a modest improvement in tolerability observed with the purl deletion alone, while achieving a high degree of specificity to the tumor microenvironment. Strains containing only the msbB' mutation, as well as strains containing the purl' and msbB' mutations, were unable to replicate, and were rapidly cleared from cells within 48 hours.

EXAMPLE 3

Salmonella asd Gene Knockout Strain Engineering and Characterization

Strain YS1646 Nisd was prepared. It is an attenuated Salmonella typhimurium strain derived from strain YS1646 (which can be purchased from ATCC, Catalog # 202165) that has been engineered to have a deletion in the asd gene. In this example, the Salmonella typhimurium strain YS1646Sasd was engineered using modifications of the method of Datsenko and Wanner (Proc. Natl. Acad. Sci. U.S.A. 97:6640-6645 (2000)), as described below.

Introduction of the Lambda Red Helper Plasmid into Strain YS1646

The YS1646 strain was prepared to be electrocompetent as described previously (Sambrook J. (1998) Molecular Cloning, A Laboratory Manual, 2 ^nd Ed., Cold Spring Harbor, NY: Cold Spring Harbor Laboratory), by growing a culture in LB and concentrating 100-fold, and then washing three times with ice-cold 10% glycerol. The electrocompetent strain was electroporated with the Lambda red helper plasmid pKD46 (SEQ ID NO:218), using a 0.2 cm gap cuvette at the following settings: 2.5 kV, 186 ohms, and 50 pF. Transformants carrying pKD46 were grown in 5 mL SOC medium with ampicillin and 1 mM L-arabinose at 30 °C, and selected on LB agar plates containing ampicillin. A YS1646 clone containing the lambda red helper plasmid pKD46 then was made electrocompetent, as described above for strain YS1646.

Construction of asd Gene Knockout Cassette

The asd gene from the genome of strain YS1646 (Broadway et al. (2014) J. Biotechnology 192:177-178) was used for designing the asd gene knockout cassette. A plasmid containing 204 and 203 base pairs (bps) of homology to the left hand and right hand regions, respectively, of the asd gene, was transformed into DH5-alpha competent cells (Thermo Fisher Scientific). A kanamycin gene cassette flanked by loxP sites was cloned into this plasmid. The asd gene knockout cassette then was PCR amplified using primers asd-1 and asd-2 (see, Table 1), and gel purified.

Deletion of asd gene

The YS1646 strain carrying plasmid pKD46 was electroporated with the gel- purified linear asd gene knock-out cassette. Electroporated cells were recovered in

SOC medium and plated onto LB agar plates supplemented with kanamycin (20 pg/mL) and diaminopimelic acid (DAP, 50 pg/mL). During this step, lambda red recombinase induces homologous recombination of the chromosomal asd gene with the kun cassette (due to the presence of homologous flanking sequences upstream and downstream of the chromosomal asd gene), and knockout of the chromosomal copy of the asd gene occurs. The presence of the disrupted asd gene in the selected kanamycin-resistant clones was confirmed by PCR amplification, with primers from the YS1646 genome flanking the sites of disruption (primer asd-3), and from the multi-cloning site (primer scFv-3) (see, Table 1). Colonies were also replica plated onto LB plates, with and without supplemental DAP, to demonstrate DAP auxotrophy. All clones with the asd gene deletion were unable to grow in the absence of supplemental DAP, demonstrating DAP auxotrophy.

Table 1. Primer Information

Kanamycin Gene Cassette Removal

The kan selectable marker was removed by using the Cre/loxP site-specific recombination system. The YS1646Aa%Z gene Kan ^R mutant was transformed with pJW168 (SEQ ID NO:224), a temperature-sensitive plasmid expressing the Cre recombinase. Amp ^R colonies were selected at 30 °C; pJW168 was subsequently eliminated by growth at 42 °C. A selected clone was tested for loss of kan by replica plating on LB agar plates with and without kanamycin, and confirmed by PCR verification using primers from the YS1646 genome flanking the sites of disruption (primers asd-3 and asd-4; for primer sequences, see Table 1).

Confirmation of Functional asd Deletion Mutant Strain YS1646A«sJ (also designated AST-101)

The asd mutant was unable to grow on LB agar plates at 37 °C, but was able to grow on LB plates containing 50 pg/mL diaminopimelic acid (DAP). The Ikasd mutant growth rate was evaluated in LB liquid media; it was unable to grow in liquid LB, but was able to grow in LB supplemented with 50 pg/mL DAP, as determined by measuring absorbance at 600 nM.

Sequence Confirmation of the asd Locus Sequence in Strain YS1646A«sd After asd Gene Deletion

The asd gene deletion strain was verified by DNA sequencing using primers asd-3 and asd-4 (see, Table 1). Sequencing of the region flanking the asd locus was performed, and the sequence confirmed that the asd gene was deleted from the YS1646 chromosome.

Complementation of asd Deletion by asd Expression from Plasmids

A plasmid, pATIU6 (SEQ ID NO:225), was chemically synthesized and assembled. The plasmid contained the following features: a high copy (pUC19) origin of replication, a U6 promoter for driving expression of a short hairpin, an ampicillin resistance gene flanked by Hindlll restriction sites for subsequent removal, and the asd gene containing 85 base pairs of sequence upstream of the start codon (SEQ ID NO:246). Into this vector, shRNAs targeting murine TREX1 were introduced by restriction digestion with Spel and Xhol, and ligation and cloning into E. coli DH5- alpha cells. The resulting plasmid was designated pATI-shTREXl.

Electroporation of Plasmids into Immunostimulatory Bacterial Strains

Selected plasmids, containing expression cassettes encoding immunostimulatory proteins and a functional asd gene, were electroporated into S. typhimurium strains lacking the asd gene with a BTX® ECM600 electroporator, using a 0.2 cm gap cuvette (BTX, San Diego, Calif.) at the following settings: 2.5 kV, 186 ohms, and 50 pF. Electroporated cells were added to 1 mL SOC supplemented with 50 pM diaminopimelic acid (DAP), incubated for 1 hour at 37 °C, and then spread onto agar plates that do not contain DAP, to select for strains that received plasmids with a functional asd gene. After single colony isolation, cell banks were produced by inoculating a flask of sterile lysogeny broth (LB) with a single well isolated colony of S. typhimurium, and incubating at 37 °C with agitation at 250 RPM. After the culture was grown to stationary phase, the bacteria were washed in PBS containing 10% glycerol, and stored in aliquots frozen at less than -60 °C.

The plasmid pATI-shTREXl was amplified in E. coli and purified for transformation into the YS1646A<25t/ strain by electroporation and clonal selection on LB Amp plates, to produce strain YS1646Aaw/-shTREXl. The YS1646Aa5<7 mutants complemented with pATIU6-derived plasmids were able to grow on LB agar and liquid media in the absence of DAP.

In a subsequent iteration, the ampicillin resistance gene (Amp ^R ) from pATI- shTREXl was replaced with a kanamycin resistance gene. This was accomplished by digestion of the pATI-shTREXl plasmid with Hindlll, followed by gel purification to remove the Amp ^R gene. The kanamycin resistance (Kan ^R ) gene was amplified by PCR using primers APR-001 and APR-002 (SEQ ID NO:226 and SEQ ID NO:227, respectively), followed by digestion with Hindlll, and ligation into the gel purified, digested pATIU6 plasmid.

In subsequent iterations, a single point mutation was introduced into the pATI-Kan plasmid at the pUC19 origin of replication, using the Q5® Site-Directed

- 347 -

RECTIFIED SHEET (RULE 91 ) ISA/EP Mutagenesis Kit (New England Biolabs) and the primers APR-003 (SEQ ID NO:228) and APR-004 (SEQ ID NO:229), to change the nucleotide T at position 148 to a C. This mutation makes the origin of replication homologous to the pBR322 origin of replication, which is a low copy origin of replication, in order to reduce the plasmid copy number.

Plasmid Maintenance Demonstrated In Vivo Using asd Complementation System

In this example, CT26 tumor-bearing mice were treated with strain YS1646 containing a plasmid that expresses an shRNA targeting TREX1 (YS1646-shTREXl), or with an a.yc/-deleted strain of YS1646, containing a plasmid with a functional asd gene and an shRNA targeting TREX1 (YS1646Aa%/-shTREXl).

CT26 (Colon Tumor #26) is a tumor model that originated from exposing BALB/c mice to N-nitro-N-methylurethane (NMU), resulting in a highly metastatic carcinoma that recapitulates the aggressive, undifferentiated and checkpointrefractory human colorectal carcinoma (see, e.g., Castle etal. (2014) BMC Genomics 15(1): 190). When implanted subcutaneously in the flank, as opposed to orthotopically in the colon, the tumor immunophenotype is much more immunosuppressive and checkpoint refractory. While largely lacking in T-cell infiltration, the tumor is rich in myeloid cells, such as macrophages and myeloid-derived suppressor cells (MDSCs) (see, e.g., Zhao et al. (2017) Oncotarget 8(33):54775-54787). As this model more closely resembles human microsatellite stable (MSS) colorectal cancer, it is an ideal model to evaluate the therapeutic approach provided herein.

For this experiment, 6-8 week-old female BALB/c mice (3 mice per group) were inoculated subcutaneously (SC) in the right flank with CT26 (purchased from ATCC) tumor cells (2xl0 ⁵ cells in 100 pL PBS). Mice bearing 8 day-old established flank tumors were intravenously (IV) injected with three doses of 5 xlO ⁶ CFUs of the YS1646Aasz/-shTREXl strain, or the parental YS1646-shTREXl strain, on days 8, 15, and 23. The plasmid encodes shTREXl as an exemplary therapeutic product; any other desired therapeutic product or products can be substituted.

Body weights and tumors were measured twice weekly. Tumor measurements were performed using electronic calipers (Fowler, Newton, MA). Tumor volume was calculated using the modified ellipsoid formula, l/2(length x width ² ). Mice were euthanized when tumor size reached >20% of body weight or became necrotic, as per IACUC regulations. At 12 days after the final Salmonella injection, tumors were homogenized, and homogenates were serially diluted and plated on LB agar plates, to enumerate the total number of colony forming units (CFUs) present, or on LB plates containing kanamycin, to enumerate the number of kanamycin resistant colonies.

The results demonstrated that 5. typhimurium strain YS1646-shTREXl did not have selective pressure to maintain the shRNA plasmid, and demonstrated significant plasmid loss, as the percent of kanamycin resistant (Kan ^R ) colonies was less than 10%. The strain that used the asd gene complementation system for plasmid maintenance, YS1646Aa«7-shTREXl, had nearly identical numbers of kanamycin resistant and kanamycin sensitive CFUs. These data demonstrate that the asd gene complementation system is sufficient to maintain the plasmid in the context of the tumor microenvironment in mice.

Enhanced Anti-Tumor Efficacy Using asd Complementation System

The asd complementation system is designed to prevent plasmid loss and potentiate the anti-tumor efficacy of the therapeutic product delivery by S. typhimurium strains in vivo. To test this, Y S 1646 Sasd strains containing the shTREXl plasmid (YS1646A«.sz/-shTREX l ), or scrambled control (YS I646At/.s<7- shSCR), that contain a functional asd gene cassette, were compared for anti-tumor efficacy in a murine colon carcinoma model, to strain YS1646 containing plasmid pEQU6-shTREXl (YS1646-shTREXl), a plasmid that lacks an asd gene cassette, and therefore, does not have a mechanism for plasmid maintenance. shTREXl is an exemplary therapeutic product.

For this experiment, 6-8 week-old female BALB/c mice (8 mice per group) were inoculated SC in the right flank with CT26 cells (2xl0 ⁵ cells in 100 pL PBS). Mice bearing established flank tumors were IV injected twice, on day 8 and on day 18, with 5xl0 ⁶ CFUs of YS1646AawZ-shTREXl, or YS1646-shTREXl, and compared to PBS control.

The YS1646-shTREXl strain demonstrated enhanced tumor control compared to PBS (70% tumor growth inhibition (TGI), day 28) despite its demonstrated plasmid loss over time. The Sasd strain containing the plasmid with the asd gene complementation system and shTREXl (YS 1646A«.sc/-shTREX I ) demonstrated superior tumor growth inhibition compared to PBS (82% TGI,/? = 0.002, day 25). These data demonstrate that improved potency is achieved by preventing plasmid loss, using the asd complementation system, and delivery of shTREXl, as compared to YS1646 containing plasmids without the asd gene complementation system. Thus, strains with asd complementation systems are superior anti-cancer therapeutics

EXAMPLE 4

S. typhinuirium Flagellin Knockout by Deletion of the fliC and fljB Genes Strain Engineering and Characterization

In the example herein, the live attenuated S. typhimurium YS1646 strain containing the asd gene deletion was further engineered to delete thc /7/C and fljB genes, in order to remove both flagellin subunits. This eliminates pro-inflammatory TLR5 activation, in order to reduce pro-inflammatory signaling and improve antitumor adaptive immunity.

Deletion of fliC Gene

In this example, fliC was deleted from the chromosome of the YS I646Ac/.st/ strain using modifications of the method of Datsenko and Wanner (Proc. Natl. Acad. Set. U.S.A. 97:6640-6645 (2000)) as described in detail in the previous example. Briefly, synthetic fliC gene homology arm sequences, that contained 224 and 245 bases of homologous sequence flanking the fliC gene, were cloned into a plasmid called pSL0147 (SEQ ID NO:230). A kanamycin gene cassette flanked by cre/loxP sites then was cloned into plasmid pSL0147, and the fliC gene knockout cassette was then PCR amplified with primers flic-1 (SEQ ID NO:232) and flic-2 (SEQ ID NO:233), gel purified, and then introduced into the YS1646Aa%7 strain carrying the temperature sensitive lambda red recombination plasmid pKD46, by electroporation. Electroporated cells were recovered in SOC + DAP medium, and plated onto LB agar plates supplemented with kanamycin (20 pg/mL) and diaminopimelic acid (DAP, 50 pg/mL). Colonies were selected and screened for insertion of the knockout fragment by PCR using primers flic-3 (SEQ ID NO:234) and flic-4 (SEQ ID NO:235). pKD46 then was cured by culturing the selected kanamycin resistant strain at 42 °C and screening for loss of ampicillin resistance. The kanamycin resistance marker then was cured by electroporation of a temperature-sensitive plasmid expressing the Cre recombinase (pJW168), and Amp ^R colonies were selected at 30 °C; pJW168 was subsequently eliminated by growing cultures at 42 °C. Selected fliC knockout clones were then tested for loss of the kanamycin marker by PCR, using primers flanking the sites of disruption (flic-3 and flic-4), and evaluation of the electrophoretic mobility on agarose gels. Deletion of fljB Gene

The fljB gene was then deleted from the Y S I 646 Ac/.s^ AfliC strain using modifications of the methods described above. Synthetic ///// gene homology arm sequences that contained 249 and 213 bases of the left hand and right hand sequence, respectively, flanking the fljB gene, were synthesized and cloned into a plasmid called pSL0148 (SEQ ID NO:231). A kanamycin gene cassette flanked by cre/loxP sites then was cloned into pSL0148, and the fljB gene knockout cassette was PCR amplified with primers fljb- 1 (SEQ ID NO:236) and fljb-2 (SEQ ID NO:237) (see, Table 1), gel purified, and introduced into strain YS I 646Amc// \fliC carrying the temperature sensitive lambda red recombination plasmid pKD46, by electroporation. The kanamycin resistance gene then was cured by Cre-mediated recombination, as described above, and the temperature-sensitive plasmids were cured by growth at non- permissive temperature. The fliC and fljB gene knockout sequences were amplified by PCR using primers flic-3 and flic-4, or fljb-3 (SEQ ID NO:238) and fljb-4 (SEQ ID NO:239), respectively, and verified by DNA sequencing. This mutant derivative of strain YS1646 was designated YSI 646A«.sz// l\fliCI ffl/B , or YS I646Az/.sz//AFLG for short.

In Vitro Characterization of Engineered S. ty phi murium Flagellin Knockout Strain

The YS1646-derived asd' mutant strain harboring the deletions of both fliC and fljB, herein referred to as YS1646Aa«7/AFLG, was evaluated for swimming motility by spotting 10 microliters of overnight cultures onto swimming plates (LB containing 0.3% agar and 50 mg/mL DAP). While motility was observed for the YS1646Az/.sz/ strain, no motility was evident with the YS1646Aa%7/AFLG strain. The YS1646AaszZ/AFLG strain then was electroporated with a plasmid containing an asd gene, and its growth rate in the absence of DAP was assessed. The YS1646Aa«//AFLG strain, with an asd complemented plasmid was able to replicate in LB in the absence of supplemental DAP, and grew at a rate comparable to the YS1646AaszZ strain containing an asd complemented plasmid. These data demonstrate that the elimination of flagellin does not decrease the fitness of 5. typhimurium in vitro.

Elimination of Flagella Decreases Pyroptosis in Murine Macrophages

5xl0 ⁵ mouse RAW macrophage cells (InvivoGen, San Diego, Ca.) were infected with the YS I646Az7.sz.//AFLG strain, or the parental YS I646Az7.sz./ strain, both harboring an asd complemented plasmid, at an MOI of approximately 100 in a gentamicin protection assay. After 24 hours of infection, culture supernatants were collected and assessed for lactate dehydrogenase release as a marker of macrophage cell death, using a Pierce™ LDH Cytotoxicity Assay Kit (Thermo Fisher Scientific, Waltham, Ma.). The YS 1646A< .sc/ strain induced 75% maximal LDH release, while the YS1646Aasz7/AFLG strain induced 54% maximal LDH release, demonstrating that deletion of the flagellin genes reduces the S. typhimurium-induced pyroptosis of infected macrophages.

Flagella-Deleted Mutants Lead to Less Pyroptosis in Infected Human Monocytes

To demonstrate that the YS1646Aa«7/AFLG strains are reduced in their ability to cause cell death in macrophages, THP-1 human macrophage cells (ATCC Catalog # 202165) were infected with the S. typhimurium strains YS1646 and YS1646AaszZ/AFLG, with the asd strain containing plasmids encoding a functional asd gene to ensure plasmid maintenance. 5xl0 ⁴ cells were placed in a 96-well dish with DMEM and 10% FBS. Cells were infected with washed log-phase cultures of S. typhimurium for 1 hour at an MOI of 100 CFUs per cell, then the cells were washed with PBS, and the media was replaced with media containing 50 pg/mL gentamicin to kill extracellular bacteria, and 50 ng/mL of IFNy to convert the monocytes into a macrophage phenotype. After 24 hours, the THP-1 cells were stained with CellTiter- Glo® reagent (Promega), and the percentage of viable cells was determined using a luminescent cell viability assay using a SpectraMax® MR plate reader (Molecular Devices) to quantify the luminescence. The cells infected with the YS1646 strain had only 38% viability, while the cells infected with the YS 1646A< .st//AFLG strain had 51% viability, indicating that the deletion of the flagellin genes induced less cell death of human macrophages, despite a very high and supraphysiological MOI.

Flagella is not Required for Tumor Colonization After Systemic Administration

To assess the impact of the flagellin knockout strains, administered in a murine model of colon carcinoma, 6-8 week-old female BALB/c mice (5 mice per group) were inoculated SC in the right flank with CT26 cells (2xl0 ⁵ cells in 100 pL PBS). Mice bearing 10-day established flank tumors were IV injected with a single dose of 3 xlO ⁵ CFUs of the YSI 646A«.sz//AFLG-shTREX I strain, or the parental YS 1646A<7.st/-shTREX l strain. At day 35 post tumor implantation, mice were euthanized, and tumors were homogenized and plated on LB plates to enumerate the number of colony forming units (CFUs) per gram of tumor tissue. The YS I646Az/.s<7- shTREXl strain colonized tumors at a mean of 5.9 xlO ⁷ CFUs per gram of tumor tissue, while the flagella-deleted YS I 646Aavt//AFEG-sliTREX I strain colonized the tumors with almost a 2-fold increased mean of 1.1 x 10 ⁸ CFUs/g of tumor tissue. The splenic colonization of the YS 1646A< .st/-shTREX I strain was calculated as a mean of 1.5 x 10 ³ CFUs/g of spleen tissue, whereas splenic colonization of the flagella-deleted YS I646A< .s/Z ^/ AFLG-shTR.EX l strain was slightly lower, at a mean of 1.2 x 10 ³ CFUs/g of spleen tissue.

These data demonstrate that the absence of flagella not only does not negatively impact tumor colonization after IV administration, but it enhances tumor colonization compared to the flagella-intact strain. Importantly, deletion of the flagella slightly reduces splenic colonization, giving a tumor to spleen ratio of 100,000-fold. These data demonstrate that, contrary to the expectation from the art, not only are the flagella not required for tumor colonization, but their elimination enhances tumor colonization, while reducing splenic colonization.

The Flagella-Deleted Strain Demonstrates Enhanced Anti-Tumor Activity in Mice

To assess the impact of the flagellin knockout strains, administered in a murine model of colon carcinoma, 6-8 week-old female BALB/c mice (5 mice per group) were inoculated SC in the right flank with CT26 cells (2xl0 ⁵ cells in 100 pL PBS). Mice bearing established flank tumors were IV injected with a single dose of 3 xlO ⁵ CFUs of the YS I646A«.sz/''AFLG-shTREX I strain, or the YS I646A«.sz/- shTREXl strain, and compared to PBS control. Mice were monitored by caliper measurements for tumor growth.

The results demonstrated that the YS I646A«.sz//AFLG-shTREX I strain, incapable of making flagella, showed enhanced tumor control compared to the parental YS I 646Ar/ -shTREX I strain (27% TGI, day 24), and significant tumor control compared to the PBS control (73% TGI, /? = 0.04, day 24). These data demonstrate that, not only is the flagella not required for tumor colonization, but its loss can enhance anti-tumor efficacy.

Flagella-Deleted Strains Demonstrate Enhanced Adaptive Immunity in a Murine Tumor Model

The impact of deletion of the flagellin on the immune response, and whether STING activation from tumor myeloid cell-delivery of shRNA to the STING checkpoint gene TREX1 would promote an adaptive type I IFN immune signature, was assessed. The CT26 murine model of colon carcinoma was used, where 6-8 week-old female BALB/c mice (5 mice per group) were inoculated SC in the right flank with CT26 cells (2xl0 ⁵ cells in 100 pL PBS). Mice bearing established flank tumors were IV injected 11 days post tumor implantation with 5 xlO ⁶ CFUs of the YS 1646Aa.scZ ^/ AFLG-shTREX l strain, or the parental YS I 646Aa.se/-shTREXl, or the scrambled plasmid control strain, YS1646Aa«7-shSCR, and compared to PBS control.

Mice were bled 7 days post-dosing on Sodium Heparin coated tubes (Becton Dickinson). Non-coagulated blood was then diluted in the same volume of PBS and peripheral blood mononuclear cells (PBMCs) were separated from the interphase layer of whole blood using Lympholyte®-M cell separation reagent (Cedarlane). Isolated PBMCs were washed with PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes at room temperature, and resuspended in flow buffer. One million PBMCs were seeded per well of a V-bottom 96-well plate. Cells were centrifuged at 1300 RPM for 3 minutes at room temperature (RT) and resuspended in 100 pL of flow buffer containing fluorochrome-conjugated AHI peptide:MHC class I tetramers (MBL International), and the cell surface flow cytometry antibodies CD4 FITC clone RM4-5; CD8a BV421 clone 53-6.7; F4/80 APC clone BM8; CD1 lb PE-Cy7 clone MI/70; CD45 BV570 clone 30-F11; CD3 PE clone 145-2C11; Ly6C BV785 clone HK1.4; I-A/I-E APC-Cy7 clone M5/114.15.2; Ly6G BV605 clone 1A8; and CD24 PercP-Cy5.5 clone MI/69 (all from BioLegend), for 45 minutes at room temperature and in the dark. After 45 minutes, the cells were washed twice with PBS + 2% FBS by centrifugation at 1200 RPM for 3 minutes. The cells were then resuspended in PBS + 2% FBS containing DAPI (4',6-diamino-2-phenylindole; dead/live stain), and data were immediately acquired using the NovoCyte® flow cytometer (ACEA Biosciences, Inc.) and analyzed using FlowJo™ software (Tree Star, Inc.).

The following cell types were enumerated as a percentage of total live cells: CD1 lb ⁺ Grl ⁺ neutrophils (possibly MDSCs, although further phenotyping in an ex vivo functional assay would be required), CD1 lb ⁺ F4/80 ⁺ macrophages, CD8 ⁺ T-cells, and CD8 ⁺ T-cells that recognize the CT26 tumor rejection antigen gp70 (AHI), the product of the envelope gene of murine leukemia virus (MuLV)-related cell surface antigen (see, e.g., Castle et al. (2014) BMC Genomics 15(1): 190).

The results, summarized in the table below, show that the YS 1646Aa.M- shSCR strain, containing a plasmid encoding a non-specific scrambled shRNA, elicits the typical anti-bacterial immune profile of significantly increased neutrophils, as compared to PBS (p = 0.02), to the flagella-intact YS 1646A<7.stZ-shTREX I strain (p = 0.02), and to the flagella-deleted strain YS1646Aasc//AFLG-shTREXl (p = 0.01), which had the lowest levels of circulating neutrophils. Similarly, bacterially-induced macrophages also were significantly elevated in the YS1646Ao«Z-shSCR strain, as compared to PBS (p = 0.01), to the YS 1646Aast/-shTREX l strain (p = 0.01), and to YS I646Ac/.sz/ ^/ AFLG-shTREX I strain (p = 0.01). Thus, both strains carrying type I IFN-inducing payloads were capable of overwriting the normal anti-bacterial immune response, which clears bacterial infections through neutrophils and macrophages, and does not induce adaptive T-cell-mediate immunity. However, while the overall circulating levels of CD8 ⁺ T-cells were similar across all groups, the flagella-deleted YS1646Aas£//AFLG-shTREXl strain demonstrated significantly increased percentages of AHl-tetramer ⁺ CD8 ⁺ T-cells, as compared to PBS (p = 0.04).

These data demonstrate the feasibility of engineering a bacteria to deliver viral-like type I IFN-inducing plasmids to tumor-resident myeloid cells. This results in a dramatic reprogramming of the immune response towards a more viral, and less bacterial, immune profile. Deletion of the flagella further enhanced the shift away from bacterially-recruited neutrophils and macrophages, and towards significantly increased tumor antigen-specific CD8 ⁺ T-cells. Thus, eliminating bacterial TLR5- mediated inflammation can enhance adaptive immunity.

SD = Standard deviation

Flagella-Deleted Strains are Restricted to the Phagocytic Myeloid Immune Cell Compartment In Vivo

According to the literature, /SfljBI (SfliC strains demonstrate suppression of many downstream genes associated with SPI-1 -mediated entry into non-phagocytic cells. In order to determine whether the YS 1646Ac/.st/AFLG strain also is deficient for non-phagocytic cell uptake, a YS I646Ac/.stZ/AFLG strain, constitutively expressing mCherry (a red fluorescent protein) under the bacterial rpsM promoter, was IV administered to MC38 subcutaneous flank tumor-bearing mice.

The MC38 (murine colon adenocarcinoma #38) model was derived similarly as the CT26 model using mutagenesis, but with dimethylhy dralazine, and in a C57BL/6 mouse strain (see, e.g., Corbett etal. (1975) Cancer Res. 35(9):2434-2439). Similarly to CT26, subcutaneous implantation results in a more T-cell excluded and immunosuppressive tumor microenvironment than when implanted orthotopically in the colon (see, e.g., Zhao et al. (2017) Oncotarget 8(33):54775-54787). MC38 has a higher mutational burden than CT26, and a similar viral-derived gp70 antigen (pl5E) that can be detected by CD8 ⁺ T-cells, although it is not considered a rejection antigen. While variants of MC38 have been found to be partially responsive to checkpoint therapy, most variants of the cell line are considered checkpoint refractory and T-cell excluded (see, e.g., Mariathasan et al. (2018) Nature 555:544-548), including the MC38 cells used herein.

For this experiment, 6-8 week-old female C57BL/6 mice (5 mice per group) were inoculated SC in the right flank with MC38 cells (5xl0 ⁵ cells in 100 pL PBS). Mice bearing large established flank tumors were IV injected on day 34 with IxlO ⁶ CFUs of the YS1646 Aa.v<7/AFLG-m Cherry strain. Tumors were resected 7 days post- IV dosing, and cut into 2-3 mm pieces into gentleMACS™ C tubes (Miltenyi Biotec) filled with 2.5 mL enzyme mix (RPMI-1640 containing 10% FBS with 1 mg/mL Collagenase IV and 20 pg/mL DNase I). The tumor pieces were dissociated using OctoMACS™ (Miltenyi Biotec) specific dissociation program (mouse implanted tumors), and the whole cell preparation was incubated with agitation for 45 minutes at 37 °C. After the 45 minute incubation, a second round of dissociation was performed using the OctoMACS™ (mouse implanted tumor) program, and the resulting single cell suspensions were filtered through a 70 pM nylon mesh into a 50 mL tube. The nylon mesh was washed once with 5 mL of RPMI-1640 + 10% FBS, and the cells were filtered a second time using a new 70 pM nylon mesh into a new 50 mL tube. The nylon mesh was washed with 5 mL of RPMI-1640 + 10% FBS, and the filtered cells were then centrifuged at 1000 RPM for 7 minutes. The resulting dissociated cells were resuspended in PBS and kept on ice before the staining process.

For the flow-cytometry staining, 100 pL of the single cell suspensions were seeded in wells of a V-bottom 96-well plate. PBS containing a dead/live stain (Zombie Aqua™, BioLegend), and Fc Blocking reagents (BD Biosciences), were added at 100 pL per well and the cells were incubated on ice for 30 minutes in the dark. After 30 minutes, the cells were washed twice with PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes. Cells were then resuspended in PBS + 2% FBS containing fluorochrome-conjugated antibodies (CD4 FITC clone RM4-5; CD8a BV421 clone 53-6.7; F4/80 APC clone BM8; CD1 lb PE-Cy7 clone MI/70; CD45 BV570 clone 30-F11; CD3 PE clone 145-2C11; Ly6C BV785 clone HK1.4; I-A/I-E APC-Cy7 clone M5/114.15.2; Ly6G BV605 clone 1 A8; and CD24 PercP-Cy5.5 clone MI/69, all from BioLegend), and incubated on ice for 30 minutes in the dark. After 30 minutes, the cells were washed twice with PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes and resuspended in flow cytometry fixation buffer (Thermo Fisher Scientific). Flow cytometry data were acquired using the NovoCyte® Flow Cytometer (ACEA Biosciences, Inc.), and analyzed using the FlowJo™ software (Tree Star, Inc ).

The results demonstrated that 7.27% of tumor-infiltrating monocytes had taken up the flagella-deleted mCherry strain in the tumor microenvironment. Similarly, 8.96% of the tumor-associated macrophage (TAM) population, and 3.33% of the tumor-infiltrating dendritic cells (DCs) had taken up the flagella-deleted mCherry strain. In contrast, within the CD45" population, corresponding to stromal and tumor cells, only 0.076% showed positivity for mCherry expression (compared to 0.067% background staining). These data demonstrate that the flagella and its downstream signaling impact on SPI-1, are necessary to enable epithelial cell infectivity, and that the lack thereof restricts uptake of the bacteria to only the phagocytic immune cell compartment of the tumor microenvironment (z.e., tumorresident immune/myeloid cells).

Deletion of the flagella confers multiple benefits to the immunostimulatory S. typhinrurium strain, including eliminating TLR5-induced inflammatory cytokines that suppress adaptive immunity, reducing macrophage pyroptosis, as well as maintaining (or enhancing) tumor-specific enrichment upon systemic administration, where uptake is confined to tumor-resident phagocytic cells. EXAMPLE 5

Salmonella pagP Gene Knockout Strain Engineering and Characterization

In this example, the YS1646Aa«Z/AFLG strain was further modified to delete pagP. The pagP gene is induced during the infectious life cycle of S. typhimurium, and encodes an enzyme (lipid A palmitoyltransferase) that modifies lipid A with palmitate. In wild-type S. typhimurium, expression of pagP results in a lipid A molecule that is hepta-acyl ated. In an msbB' mutant, in which the terminal acyl chain of lipid A cannot be added, the expression of pagP results in a hexa-acylated lipid A molecule. LPS with hexa-acylated lipid A has been shown to be highly pro- inflammatory, and to have a high affinity for TLR4 (hepta-acylated lipid A, found in wild-type, has the highest affinity for TLR4). In this example, a strain deleted of pagP and msbB can produce only penta-acyl ated lipid A, allowing for lower pro- inflammatory cytokines due to low affinity for TLR4, enhanced tolerability, and increased adaptive immunity when the bacteria are engineered to deliver plasmids encoding immunomodulatory proteins.

SpagP Strain Construction

The pagP gene was deleted from the YS1646Aast//AFLG strain using modifications of the methods described in the preceding examples. Synthetic pagP gene homology arm sequences that contain 203 and 279 bases of the left hand and right hand sequence, respectively, flanking the pagP gene, were synthesized and cloned into a plasmid called pSL0191 (SEQ ID NO:331). A kanamycin gene cassette flanked by cre/loxP sites then was cloned into pSL0191, and the pagP gene knockout cassette was PCR amplified with primers pagp-1 (SEQ ID NO:315) and pagp-2 (SEQ ID NO:316) (see, Table 1), gel purified, and introduced into strain YS1646AaszZ/AFLG, carrying the temperature sensitive lambda red recombination plasmid pKD46, by electroporation. The kanamycin resistance gene then was cured by Cre-mediated recombination, as described above, and the temperature-sensitive plasmids were cured by growth at non-permissive temperature. The pagP gene knockout sequences were amplified by PCR using primers pagp-3 (SEQ ID NO:317) and pagp-4 (SEQ ID NOG 18), and verified by DNA sequencing. The resulting mutant derivative of YS1646 was designated 6 6Sasdl AFLG/ SpagP . pagP Deletion Mutants have LPS with Penta-Acylated Lipid A, and Induce Reduced Inflammatory Cytokines The pagP gene also was deleted from the YS I646Aa.s<7 strain using the lambda-derived Red recombination system, as described in Datsenko and Wanner (Proc. Natl. Acad. Sci. U.S.A. 97:6640-6645 (2000)) and above, to generate the strain NS(646Sasd! SpagP . This strain was then electroporated with a plasmid containing a functional asd gene, to complement the deleted asd gene and to ensure plasmid maintenance in vivo. The lipid A was then extracted from this strain and evaluated by Matrix- Assisted Laser Desorption/Ionization Mass Spectrometry (MALDI MS) and compared to lipid A from the wild-type 5. typhimurium strain ATCC 14028, the YS1646 strain (which is deleted for msbB and purl), and the YS 1646Aasc/ strain. Wild-type Salmonella had a minor lipid A peak with a mass of 2034, and a major peak with a mass of 1796, corresponding to the hepta-acylated and hexa-acylated species, respectively, due to the presence of a functional msbB gene. The msbB- deleted strains, YS1646 and YS1646AaszZ, had major peaks at 1828 and 1585, corresponding to a mixture of hexa-acylated and penta-acylated lipid A. The msbB and pagP deleted strain, NS(646SasdlNpagP, had only a single peak with a mass of 1585, corresponding to penta-acylated lipid A. These data demonstrate that deletion of pagP prevents palmitoylation of the lipid A, thereby restricting it to a single penta- acylated species.

To determine whether the LPS with penta-acylated lipid A from the SpagP mutant strains reduced TLR4 signaling, 4 pg of purified LPS from the wild-type strain, the YS1646 strain, or the YS I646Aa.v// ApagP strain, was added to THP-1 human monocytic cells (ATCC Catalog # TIB-202), and the supernatants were evaluated 24 hours later for the presence of inflammatory cytokines using a Cytometric Bead Array (CBA) kit (BD Biosciences). The results showed that LPS from the NS\646Sasdl NpagP strain induced 25% of the amount of TNFa, compared to wild-type LPS, and induced 7-fold less IL-6 than wild-type LPS. The LPS from the YS1646Sasd/SpagP strain induced 22-fold less IL-6 than strain YS1646, demonstrating that the penta-acylated LPS species from a SpagP mutant is significantly less inflammatory in human cells, and indicating that the NpagP mutant would be better tolerated in humans. Deletion of pagP Induces Significantly Less IL-6 in Primary Human M2 Macrophages

To demonstrate that the YS1646Aasfi//AFLG/Apo' strain also elicits less inflammatory and dose-limiting IL-6 from primary human M2 macrophages, the strain was evaluated, and compared with the YS1646Aa%Z/AFLG and the parental YS1646 strains. The M2 macrophages derived from human donors are representative of the immunosuppressive phenotypes that are highly enriched in T-cell excluded solid tumors. Frozen human PBMCs, isolated from healthy human donors, were thawed in complete medium (RPMI-1640 + IX non-essential amino acids + 5% human AB serum), and washed by centrifugation for 10 minutes at 800 RPM at room temperature. PBMCs were resuspended in PBS + 2% FBS, and monocytes were negatively isolated using a CD 16 depletion kit (StemCell Technologies). Isolated untouched monocytes were then washed by centrifugation in PBS + 2% FBS and resuspended in complete medium containing 100 ng/mL human macrophage colonystimulating factor (M-CSF) and 10 ng/mL human IL-4. Isolated monocytes (3e5 per well) were then seeded in a 24-well plate with a final volume of 750 pL. Two days after seeding, the cell culture media was entirely aspirated and replaced with fresh complete medium containing 100 ng/mL human M-CSF and 10 ng/mL human IL-4. Two days later (on day 4), 500 pL of complete medium, containing cytokines 100 ng/mL human M-CSF and 10 ng/mL human IL-4, was added per well for 48 hours. On day 6, the cell culture media was entirely aspirated and replaced with fresh complete medium without cytokines, alone, or with media containing the log-phase cultures of the S. typhimurium strains at an MOI of 20. Cells were infected for 1 hour, then washed with PBS, and the media was replaced with fresh media containing 50 pg/mL gentamicin to kill extracellular bacteria. The wells were then washed and replaced with fresh media and allowed to incubate at 37 °C and 5% CO2. After 48 hours, supernatants were harvested and assayed for cytokines using a human IL-6 cytometric bead array (CB A) kit (BD Biosciences), according to the manufacturer’ s instructions.

The results demonstrated that secreted IL-6 levels from human primary M2 macrophages, infected with parental strain YS1646, yielded an average of 14839 ± 926 pg/mL, while the IL-6 levels from the YS I 646A«.sz//AFLG strain were significantly lower, at 2075 ± 723 pg/mL (p = 0.004). This further affirms the impact that the deletion of flagella, and elimination of TLR5 signaling, has on the induction of IL-6. The YS 1646Aast//AFLG/ApagP strain elicited the lowest IL-6 levels, at 332 ± 100 pg/mL, demonstrating the reduced ability of this modified LPS coating to stimulate TLR4, and the resulting dramatically reduced inflammatory IL-6 production.

The Combined Flagella and pagP Deletions Significantly Enhance Tolerability in Mice

To determine whether the modified strains described above are more attenuated than parental strain YS1646, a median lethal dose (LD50) study was conducted. 6-8 week-old BALB/c mice (5 mice per group) were injected intravenously with a dose range of 3e5 to 3e7 CFUs of strain YS1646, or the derivative strains YS1646Ac7.st//AFLG, YS1646Aasz// hpagP, and YS1646Aasz//AFLG/A/iu'gP. Unlike strain YS1646, the derivative strains also carried a plasmid encoding murine IL-2, an FDA-approved cytokine that has demonstrated significant toxicity when systemically administered.

The LD50 for strain YS1646 was found to be 4.4 x 10 ⁶ CFUs (average of two studies), in line with previously published LD50 reports of YS1646, and a >1000-fold improvement compared to wild-type S. typhimurium (see, e.g, Clairmont et al. (2000) J. Infect. Dis. 181:1996-2002). The LD50 for the YS I 646 Ac7.st//AFLG strain was determined to be 2.07 x 10 ⁷ CFUs, demonstrating a greater than 4.5-fold reduction in virulence compared to strain YS1646. The LD50 for the YS1646 Aasd/ ApagP strain was determined to be 1.39 x 10 ⁵ CFUs, demonstrating at least a 3.2-fold reduction in virulence compared to strain YS1646, which is expected, given that the strain still has highly inflammatory flagella. The LD50 for the YS I646Ai7.sz.//AFLG/A/v/gP strain could not be established, as no mice died at the highest dose given, but was >6.2 x 10 ⁷ CFUs. The YS1646Aast7/AFLG/A/?agP strain therefore demonstrates a >14-fold reduction in virulence compared to parental YS1646 strain. These data demonstrate that the genetic modifications described above reduce the virulence of the clinical S. typhimurium strain, YS1646 (also known as VNP20009), and therefore, lead to increased tolerability in humans.

In the Phase I clinical trial of VNP20009 (see, e.g, Toso et al. (2002) J. Clin. Oncol. 20(1): 142-152), the presence of the bacteria in patients’ tumors only partially was observed at the two highest doses tested, 3 x 10 ⁸ CFU/m ² (33% presence), and 1 x 10 ⁹ CFU/m ² (50% presence), indicating that the tolerable dose of VNP20009 was too low to achieve tumor colonization. By improving the tolerability of the strains through the modifications described above, >14-fold higher doses can be administered, if necessary, improving the percentage of patients whose tumors will be colonized, and increasing the level of therapeutic colonization per tumor, thereby solving the observed problems with VNP20009.

The Combined Flagella and pagP Deletions Significantly Limit the Generation of Anti-5. typhimurium Antibodies in Mice

The surviving mice from the 3 x 10 ⁶ CFU dosing group described above (N=5, except for N=4 in the YS1646 dosing group) were kept for 40 days post IV dosing, at which time they were bled for serum, and assessed for antibody titers to 5. typhimurium, by a modified flow-based antibody titering system. Overnight cultures of the YS I646Ar/.st//AFLG-mCherry strain were washed and fixed with flow cytometry fixation buffer. Sera from the previously-treated mice, and from naive control mice, were seeded in a 96-well plate, and serial dilutions were performed in PBS. Next, 25 pL of the YS I646A«.vt//AFLG-mCherry cultures, containing 1 x 10 ⁶ CFUs, were added to the sera and incubated for 25 minutes at room temperature (RT). The bacteria were then washed twice with PBS by spinning them at 4000 RPM for 5 minutes. After the last wash, the bacteria were resuspended in PBS containing a secondary Goat anti-Mouse Fc AF488 antibody (1/400 dilution from stock), and incubated for 25 minutes at room temperature and protected from light. The bacteria were then washed three times with PBS by spinning them at 4000 RPM for 5 minutes. After the last wash, the bacteria were resuspended in PBS and data were acquired using the NovoCyte® flow cytometer (ACEA Biosciences, Inc.), and analyzed using the MFI FlowJo™ software (Tree Star, Inc.).

To evaluate the results by flow cytometry, the highest dilution with signal in all groups was chosen (the 1250X serum dilution), and the corresponding mean fluorescence intensity (MFI) values were plotted. The limit of detection (LOD) was chosen at an MFI of 1000, as that is the MFI obtained without staining, as well as with background staining with Goat anti-Mouse Fc AF488 antibody only. Therefore, an MFI greater than 1000 was considered a positive signal, and everything equal to or under this value was considered a negative result, despite having an MFI value.

The results of this assay revealed a high MFI titer of anti-5, typhimurium serum antibodies from mice treated with 3 x 10 ⁶ CFUs of the YS1646 strain (MFI of 29196.3 ± 20730), in line with previously published data that YS1646 is able to generate serum antibodies (that are non-neutralizing). Fewer antibodies were detected in the mice treated with the YS1646 Aasz// AFLG strain (MFI of 11257 ± 9290), which can be due to the lack of adjuvant activity from the flagella. In the mice treated with the S1646 Aasd/ pagP strain, significantly fewer antibodies were generated (MFI of 4494 ± 3861), as compared to strain YS1646 (p = 0.033), which can be due to the altered LPS surface coating. The most significant reduction in serum antibodies was demonstrated in the YS1646 Easdl AFLG/ pagP treatment group (MFI of 1930 ± 2445), where several of the mice had MFI titers under 1000, and were thus considered negative for serum antibodies (p = 0.021, vs. strain YS1646). Thus, the combined deletions of the flagella and the pagP gene enable both improved safety, as well as significantly reduced immunogenicity, which will enable repeat dosing of high CFUs in humans. pagP and Flagella Deleted Strains, and their Combination, Demonstrate Significantly Higher Viability in Human Serum Compared to Strain YS1646

Strain YS1646 exhibits limited tumor colonization in humans after systemic administration. It is shown herein that strain YS1646 is inactivated by complement factors in human blood. To demonstrate this, strains YS1646 and E. coli DI 0B were compared to exemplary immunostimulatory bacteria provided herein, that contain additional mutations that alter the surface of the bacteria. These exemplary modified strains were YS I 646Ac/.s<7/ EpagP , YS1646 asd! AFLG, and YS1646Aa%AAFLG/A/?u' F ^> . These three strains, in addition to YS1646 and E. coli DI 0B cultures, were incubated with serum, or with heat-inactivated (HI) serum, from either pooled mouse blood, or pooled healthy human donors (n=3), for 3 hours at 37 °C. After incubation with serum, bacteria were serially diluted and plated on LB agar plates, and the colony forming units (CFUs) were determined.

In mouse serum, all strains remained 100% viable and were completely resistant to complement inactivation. In human serum, all strains were 100% viable in the heat-inactivated serum. The A. coli DI 0B strain was completely eliminated after 3 hours in whole human serum. In whole human serum, the YS1646 strain exhibited only 6.37% of live colonies, demonstrating that tumor colonization of the YS1646 clinical strain was limited due to complement inactivation in human blood. For the YS1646 asd! AFLG strain, 31.47% of live colonies remained, and for the YS1646 Aasa/ApagP strain, 72.9% of live colonies remained, after incubation with human serum for 3 hours. The combined YS1646Aa%//AFLG/A/?a'gP strain was completely resistant to complement in human serum.

These data explain why strain YS1646 (VNP20009) has very low tumor colonization when systemically administered. It is shown herein that strain YS1646 is highly sensitive to complement inactivation in human serum, but not in mouse serum. These data explain why limited tumor colonization was observed in humans, while mouse tumors were colonized at a high level. The fljBlfliC or pagP deletions, or the combination of these mutations, partially or completely rescues this phenotype. Thus, the enhanced stability observed in human serum with the YS I 646Aa.S6//A/x7g/ ^? , YS 1646A<7.st/ ^/ AFLG, and YS I 646A«.%//AFLG/A/;<:^/ ^? strains provides for increased human tumor colonization.

EXAMPLE 6

Salmonella ansB Gene Knockout Strain Engineering and Characterization

In this example, the YS I 646Aa.sz//AFLG/A/;a ₍ q/ ^J strain was further modified to delete ansB, the gene encoding bacterial L-asparaginase II. Secretion of L- asparaginase II by S. typhimurium in the presence of T-cells has been shown to directly impair T-cell function, by reducing T-cell receptor (TCR) expression, and impairing cytolytic cytokine production. As a result, bacteri ally-derived asparaginases have been successfully used to treat acute lymphoblastic leukemia (ALL) for decades. Deletion of ansB from the bacterial genome eliminates the ability of the S. typhimurium strain to produce L-asparaginase II, thereby enhancing the function of T- cells in the bacterially-colonized tumor microenvironment.

SansB Strain Construction

The ansB gene was deleted from the YS I646Aas///AFLG/A/%/g/ ^> strain using modifications of the methods described in the preceding examples. Synthetic ansB gene homology arm sequences that contained 236 and 251 bases of the left hand and right hand sequence, respectively, flanking the ansB gene, were synthesized and cloned into a plasmid called pSL0230 (SEQ ID NO:332). A kanamycin gene cassette flanked by cre/loxP sites then was cloned into plasmid pSL0230, and the ansB gene knockout cassette was PCR amplified with primers ansb-1 (SEQ ID NO: 319) and ansb-2 (SEQ ID NO: 320), gel purified, and introduced into strain YS1646Aa%//AFLG/A/rag7 ^? , carrying the temperature sensitive lambda red recombination plasmid pKD46, by electroporation. The kanamycin resistance gene then was cured by Cre-mediated recombination, as described above, and the temperature-sensitive plasmids were cured by growth at non-permissive temperature. The ansB gene knockout sequences were amplified by PCR using primers ansb-3 (SEQ ID NO:321) and ansb-4 (SEQ ID NO:322) (see, Table 1), and verified by DNA sequencing. The resulting mutant derivative of YS 1646 was designated YS 1646 Xasd! AFLG/A/ragT’/Aczns.S.

Deletion of ansB Eliminates Asparaginase Activity In Vitro

In order to determine whether the YS \ 646Xasd/XFLG/XpagP/Xa/isB strain produced less L-asparaginase II, cultures of this strain, along with the ansB-miacl YS1646Arz56?/AFLG/AprzgF ^> strain, were grown in LB and allowed to reach stationary phase. At this time, 50 pL of conditioned media from the cultures was analyzed for asparaginase activity, using a colorimetric Asparaginase assay kit (Sigma-Aldrich), per the manufacturer’s instructions. After a 40 minute incubation, the absorbance units were read on a SpectraMax® M3 Spectrophotometer (Molecular Devices) at an absorbance wavelength of 570 nm.

Compared to the recombinant L-asparaginase II positive control, which gave an absorbance of 1.95, the absorbance of the ansB -intact YS1646 Aa«// AFLG/ AptzgP strain was 0.82. Deletion of ansB, in the YS1646A«.S6// AFLG/ XpagPI XansB strain, however, resulted in background levels of asparaginase activity, detected at an absorbance of 0.109. These data confirm that the XansB mutation completely eliminates asparaginase activity.

Deletion of ansB Restores T-Cell Function in an In Vitro Co-Culture Assay

In order to functionally characterize the ansB-deleted strain for the impact of reduced L-asparaginase II activity on T-cells, a co-culture assay was established using strain-infected murine primary bone marrow-derived macrophages (BMMs) in culture with splenic purified T-cells. For this assay, spleens from healthy BALB/c mice were isolated and dissociated, and splenic CD4 ⁺ and CD8 ⁺ T-cells were isolated using a mouse T-cell isolation kit (StemCell Technologies), per the manufacturer’s instructions. From the isolated T-cells, 2e5 cells were added per well to a Flat-bottom 96-well plate that had been previously coated with 5 pg/ml of an anti-mouse CD3s antibody (clone 145-2C11, Thermo Fisher Scientific). Conditioned LB media from the YS1646 Xasdi AFLG/ XpagP and YS I 646Aast// AFLG/ XpagPI XansB cultures, grown to stationary phase, were filtered through a 0.45 pM nylon mesh and added to the T-cells, with or without the addition of 10 ug/ml of an agonistic anti-CD28 antibody for co-stimulation. Control groups containing recombinant asparaginase at 20 U/mL, and normal culture media, were used as controls in the assay. The plate was incubated at 37 °C in a 5% CO2 incubator. At 24 hours post-incubation, 100 pL of the co-culture supernatants were harvested from the wells, and a murine Thl-specific cytokine bead array (CBA, BioLegend) was performed. Concurrently, T-cells were harvested and analyzed for surface T-cell receptor p (TCRp) expression on CD4 ⁺ and CD8 ⁺ T-cells by flow cytometry, as well as for intracellular staining of IFNy, TNFa, and IL-2.

The results confirmed that the ansB-intact strain (YS1646A«.sc//AFLG/A/?67 _< g/-’) used to infect the macrophages, and subsequently co-cultured with T-cells, induces profound T-cell immunosuppression. This was exhibited by marked downregulation of TCRp surface expression in both CD4 ⁺ and CD8 ⁺ T-cells (see table below), compared to media control, and to the positive control of recombinant asparaginase at a concentration of 20 U/mL. Deletion of ansB in the YS1646Aa«Z/AFLG/A/iagP/Aaft£S strain significantly restored TCRp surface expression in both CD4 ⁺ (p = 0.004) and CD8 ⁺ T-cells (p = 0.002), as compared to the parental YS1646 Aast// AFLG/ A/ragP strain.

SD = Standard deviation

T-cell secretion of cytokines, 24 hours after co-culture, was measured as a marker of T-cell cytolytic function. As shown in the table below, T-cell production of the cytokines IFNy, TNFa, and IL-2 was markedly lower after treatment with the YS I646Aa.st/ ^/ AFLG/A/?a _< g/ ^> strain, as compared to the media control, and was significantly restored by ansB deletion in the YS 1646Aa.sc// AFLG/ hpagPI AansB strain (fFNy (p = 0.05), TNFa (p = 0.012), and IL-2 (p = 0.006)). These data indicate that deletion of ansB in the YS I 646Aa.sc// AFLG/ kpagP/ l\ansB strain significantly restores T-cell cytolytic function, as compared to the parental YS1646Aa.sc/ ^/ AFLG/A/?a _< g/ ^> strain.

SD = Standard deviation

Deletion of ansB Restores Tumor-Resident T-Cell TCRp Expression In Vivo

In in vitro co-culture assays, expression of ansB demonstrated immunosuppressive effects on T-cell function, including downregulation of TCRP on T-cells by flow cytometry. In order to assess whether this would similarly occur in vivo, the MC38 mouse model of colorectal cancer was utilized.

For this experiment, 6-8 week-old female C57BL/6 mice (4 mice per group) were inoculated SC in the right flank with MC38 cells (5xl0 ⁵ cells in 100 pL PBS). Mice bearing large established flank tumors were IV injected on day 17 with IxlO ⁷ CFUs of the YS1646 Ar/.s<7/AFLG-m Cherry strain. Tumors were resected 7 days post IV dosing, and cut into 2-3 mm pieces into gentleMACS™ C tubes (Miltenyi Biotec) filled with 2.5 m enzyme mix (RPMI-1640 containing 10% FBS with 1 mg/mL Collagenase IV and 20 pg/mL DNase I). The tumor pieces were dissociated using OctoMACS™ (Miltenyi Biotec) specific dissociation program (mouse implanted tumors), and the whole cell preparation was incubated with agitation for 45 minutes at 37 °C. After 45 minutes of incubation, a second round of dissociation was performed using the OctoMACS™ (mouse implanted tumor) program, and the resulting single cell suspensions were filtered through a 70 pM nylon mesh into a 50 mL tube. The nylon mesh was washed once with 5 mL of RPMI-1640 + 10% FBS, and the cells were filtered a second time using a new 70 pM nylon mesh, into a new 50 mL tube. The nylon mesh was washed with 5 mL of RPMI-1640 + 10% FBS, and the filtered cells were then centrifuged at 1000 RPM for 7 minutes. The resulting dissociated cells were resuspended in PBS and kept on ice before the staining process.

For the flow-cytometry staining, 100 pL of the single cell suspensions were seeded in wells of a V-bottom 96-well plate. PBS containing a dead/live stain (Zombie Aqua™, BioLegend) and Fc Blocking reagents (BD Biosciences) were added at 100 pL per well, and the cells were incubated on ice for 30 minutes in the dark. After 30 minutes, the cells were washed twice with PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes. Cells were then resuspended in PBS + 2% FBS containing fluorochrome-conjugated antibodies (CD45 BV570 clone 30-F11; TCRp PE clone H57-597; and CD4 FITC clone RM4-5; all from BioLegend) and DAPI (BioLegend), and incubated on ice for 30 minutes in the dark. After 30 minutes, the cells were washed twice with PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes, and resuspended in flow cytometry fixation buffer (Thermo Fisher Scientific). Flow cytometry data were acquired using the ACEA NovoCyte® flow cytometer (ACEA Biosciences, Inc.), and analyzed using the FlowJo™ software (Tree Star, Inc.).

As shown in the table below, the average mean fluorescence intensity (MFI) for the surface expression of TCRp on tumor-infiltrating CD4 ⁺ T-cells, following their interaction within tumors with the colonized parental YS1646 Aavt// AFLG/ A/?czgP strain, was significantly lower than with the c///.s7Ldeleted

YS1646Aasz// AFLG/ SpagPI SansB strain (p = 0.042), which was higher than even the PBS control -treated mice.

MFI = mean fluorescence intensity; AVG = average; SD = standard deviation

Taken together, these data confirm the necessity of deleting the ansB gene in order to restore T-cell function, due to the bacterial production of immunosuppressive L-asparaginase II, and demonstrate the enhanced T-cell function observed with the ansB deletion in the YS I 646Aa.s<7/ AFLG/ SpagP 7 SansB strain.

EXAMPLE 7

Salmonella csgD Gene Knockout Strain Engineering and Characterization

The YS I646Ac/.sz// AFLG/ SpagPI SansB strain was further modified to delete csgD, a master gene that controls 5. typhimurium curli fimbriae formation, cellulose production, and c-di-GMP production. The csgD gene deletion eliminates the possibility of cellulose-mediated biofilm formation, reduces pro-inflammatory signaling, and enhances uptake by host phagocytic cells. This increase in intracellular localization would thereby enhance the effectiveness of plasmid delivery and immunomodulatory protein production.

Xcsgl) Strain Construction

The csgD gene was deleted from the YS I 646Ac/.s<7/ AFLG/ XpagPI XansB strain, using modifications of the methods described in the preceding examples. Synthetic csgD gene homology arm sequences that contained 207 and 209 bases of the left hand and right hand sequence, respectively, flanking the csgD gene, were synthesized and cloned into a plasmid called pSL0196 (SEQ ID NO:333). A kanamycin gene cassette flanked by cre/loxP sites then was cloned into plasmid pSL0196, and the csgD gene knockout cassette was PCR amplified with primers csgd-1 (SEQ ID NO: 323) and csgd-2 (SEQ ID NO: 324), gel purified, and introduced into strain YS1646 Xcisd/ AFLG/ XpagP/ XansB, carrying the temperature sensitive lambda red recombination plasmid pKD46, by electroporation. The kanamycin resistance gene then was cured by Cre-mediated recombination, as described above, and the temperature-sensitive plasmids were cured by growth at non-permissive temperature. The csgD gene knockout sequences were amplified by PCR, using primers csgd-3 (SEQ ID NO:325) and csgd-4 (SEQ ID NO:326), and verified by DNA sequencing. The resulting mutant derivative of parental strain YS1646 was designated YS 1646 Xcisdi AFLG/ XpagPI XcmsBi Xcsgl). csgD-Deleted Strains Cannot Form RD AR Colonies on Congo Red Plates

The ability to form Rough Dry And Red (RD AR) colonies after growth on Congo Red plates is a well-validated assay for bacterial biofilm formation. The Rough and Dry texture occurs through cellulose production, and the red is due to the accumulation of pigment by the curli fimbriae surface structures. For this assay, the YS1646Aa«Z/AFLG/A/iagP/Ai»7£S strain was compared to the YS I bdbAc/.st// AFLG/ XpagP! XansB! XcsgD strain for the ability to form the RD AR phenotype after incubation on Congo Red agar plates.

Congo Red agar plates were prepared with soytone (10 g/L) and yeast extract (5 g/L) (modified LB without NaCl), and complemented with Congo red (40 mg/L) and Coomassie brilliant blue G-250 (20 mg/L). Five microliters of a stationary phase bacterial culture was spotted onto Congo Red plates, and incubated at 37 °C for 16 hours, then transferred to 30 °C and incubated for an additional 120 hours. Visual analysis of colony morphology and color was performed and recorded daily to confirm presence or absence of the RD AR colony morphotype. Comparing the colony morphotypes between the two strains, the YS1646Aa%Z/AFLG/A/ifl'gP/A£OTs5/AcsgD strain had a smooth phenotype, and the colonies lacked pigment. In comparison, the YS I 646Aav6//AFLCr/A/?«g/7Acm.s/i strain, still containing the csgD gene, exhibited the classic rough and dry appearance, and clear evidence of pigment uptake. Thus, the functional assay confirms that the EcsgD strain is unable to form biofilms, as it lacks curli fimbriae and cellulose production. csgD- Deleted Strains Demonstrate Superior Anti-Tumor Efficacy in a Highly Refractory Mouse Model of Triple Negative Breast Cancer

The impact of the csgD deletion in models where the immunostimulatory bacterial therapy colonizes tumors, but has shown limited efficacy, was assessed. This can indicate the presence of bacterially-produced cellulose that can limit uptake into tumor-resident myeloid cells, thereby limiting therapeutic benefit (see, e.g., Crull et al. (2011) Cellular Microbiology 13(8): 1223-1233). The difficult-to-treat EMT6 model was utilized, which is a representative model of human triple negative breast cancer (see, e.g., Yu et al. (2018) PLoS ONE 13(1 l):e0206223). When EMT6 tumor cells are administered orthotopically into the mammary fat pad, as opposed to subcutaneously in the flank, the model is T-cell excluded, highly metastatic, and highly refractory to immunotherapy, including to all approved checkpoint antibodies (see, e.g., Mariathasan et al. (2018) Nature 554:544-548).

For this experiment, 6-8 week-old female BALB/c mice (5 mice per group) were inoculated in the left mammary fat pad with EMT6 tumor cells (ATCC # CRL- 2755) (2xl0 ⁵ cells in 100 pL PBS). Mice bearing 13 day-old established mammary tumors (~55 mm ³ in volume) were IV injected with a single dose of 1 xlO ⁷ CFUs of the c.sg/J-deleted YS I646Aa.st// NFLG/ApagPl EansBi EcsgD strain, or the parental YS IbdbAc/.st// AFLG/ NpagPI EansB strain, and compared to PBS control. The bacterial strains contained a plasmid encoding a constitutively active murine STING (EF-la muSTING R283G).

The tumors in the PBS-treated mice grew evenly, reaching a max tumor volume at day 35 (1199.0 ± 298.1 mm ³ ). Mice treated with the csgD-intact strain, YS1646 Easdl AFLG/ NpagPI EansB, did not demonstrate evidence of anti-tumor efficacy in this model, also reaching max tumor volume at day 35 (1689.1 ± 537.0). Ex vivo LB plating of these tumors revealed all tumors to be colonized. However, the c.s /Ldeleted strain, YS1646 Kasai AFLG/ KpagPI KansBI KcsgD, resulted in 3 out of 5 mice being completely cured of both their primary and any metastatic disease (day 60+). Overall tumor growth inhibition (TGI) was 45.7%, with one of the other two remaining tumors partially responding before eventually growing out. The two bacterial strains contained the same plasmid payload, yet only one demonstrated significant anti-tumor efficacy. Thus, in one of the most intractable and highly metastatic syngeneic tumor models, orthotopic EMT6, a strain with a csgD deletion was able to induce systemic anti-tumor efficacy, and result in 60% complete responses. csgD-Deleted Strains Demonstrate Enhanced Intracellular Uptake In Vivo

In order to determine whether the c gD-deleted strain demonstrated improved efficacy because of greater bacterial uptake into tumor-resident myeloid cells, an ex vivo gentamicin protection assay was performed (see, e.g., Crull el al. (2011) Cellular Microbiology 13(8): 1223-1233). For this experiment, 6-8 week-old female C57BL/6 mice (4 mice per group) were inoculated SC in the right flank with MC38 cells (5xl0 ⁵ cells in 100 nL PBS). Mice bearing large established flank tumors were IV injected on day 17 with IxlO ⁷ CFUs of the c D-deleted YS1646Aa«// AFLG/ KpagPI KansB! Kcsgl) strain (N=12), or the parental YS1646 strain (N=4). Tumors were resected 7 days post IV dosing, weighed, and minced in RPMI supplemented with 1 mg/mL collagenase IV and 20 mg/mL DNase I, and incubated with shaking at 37 °C for 30 minutes, to generate a single cell suspension. After 30 minutes, the suspension was passed through a 70 mm filter, and the recovered volume was divided into two separate, identical samples. Gentamicin (Thermo Fisher Scientific) was added at a concentration of 200 mg/mL to one of each of the paired samples to kill extracellular bacteria, and the samples were incubated with shaking at 37 °C for 90 minutes. Cell suspension samples were then washed and lysed with 0.05% Triton X, and plated on LB agar plates to enumerate for CFUs.

The results demonstrate that, compared to the CFUs from YS1646-treated tumors without gentamicin treatment (11925 ± 19859 CFUs), gentamicin treatment resulted in very few CFUs detected from the tumors (51 ± 45 CFUs). This indicates that the bacteria reside largely extracellularly in these tumors, and are thus sensitive to gentamicin elimination. In the c D-deleted

YS I646Ac/.s<7 ^/ AFLG/ KpagPI KansBI KcsgD treatment group, the non-gentamicin treated tumors yielded high CFUs, as expected from well-colonized tumors, and treatment with gentamicin yielded less CFUs (1276 ± 2410 CFUs), and much more than in the parental YS1646 strain-treated tumors. This is due to more of the csgD- deleted bacteria residing intracellularly, and thus, being protected from gentamicin. These data demonstrate that the csgD deletion improves intracellular uptake of the bacteria, which can enhance plasmid delivery of immunomodulatory proteins in vivo.

EXAMPLE 8 pATI-1.75 and pATI-1.76 Vector Construction

A plasmid was designed and synthesized that contains the following features: a pBR322 origin of replication, the asd gene, a kanamycin resistance gene flanked by Hindlll sites for curing, and a multiple cloning site for expression cassette insertion. The expression cassette is composed of multiple elements, including eukaryotic promoters, open reading frames (ORFs), post-transcriptional regulatory elements, and polyadenylation signals, that are assembled in various configurations.

Exemplary promoters include the human cytomegalovirus (CMV) immediate early core promoter encoded directly downstream of the CMV immediate early enhancer sequence, and the core promoter for human elongation factor-1 alpha (EFla). Open reading frames (ORFs) can include one or more sequences that each are translated into a protein, and can be separated into distinct polypeptides by insertion of a 2A sequence, whereby eukaryotic ribosomes fail to insert a peptide bond between Gly and Pro residues within the 2A sequence. Examples of 2A sequences are the T2A peptide (see, e.g., SEQ ID NO:327) from the Thosea asigna virus (TaV) capsid protein, and the P2A peptide (see, e.g., SEQ ID NO:328) from porcine teschovirus (PTV). Upstream furin cleavage sites (RRKR), and other enhancer elements, are placed upstream to facilitate cleavage to separate the expressed proteins.

Examples of post-transcriptional regulatory elements (PREs) include the Woodchuck Hepatitis virus PRE (WPRE; SEQ ID NO:346) and the Hepatitis B virus PRE (HPRE; SEQ ID NO:347), which increase accumulation of the cytoplasmic mRNA of a gene by promoting mRNA nuclear export to the cytoplasm, enhancing 3’ end processing and stability. Examples of polyadenylation signal sequences include the SV40 polyadenylation (SV40polyA, or SV40pA) signal, and the bovine growth hormone polyadenylation (bGHpolyA, or bGHpA) signal, both of which are 3’ regulatory elements that serve to promote transcriptional termination, and contain the sequence motif recognized by the RNA cleavage complex.

Figures 1A-C depict the pATI-1.75 and pATI-1 76 (Fig. 1A and IB, respectively) plasmid inserts; Figure 1C shows that the Shine-Dalgamo sequence is replaced with a Kozak sequence for translation in eukaryotic cells, such as myeloid cells. Plasmids with such sequences and constructions can be used in embodiments in which the bacteria are designed to deliver RNA, which cannot be translated by the bacterial host, but can be translated in vivo in a eukaryote.

EXAMPLE 9

Designed Heterologous Protein Expression Plasmids Induce Functional Protein Production from Human Cells

Optimal Expression of Cytokines Established in Human Cells

In order to exemplify that immunostimulatory cytokines can be expressed from designed plasmids in human cells, a panel of cytokines were cloned into the pATI-1.75 plasmid under the control of the EF-la promoter. The cytokines include, but are not limited to, murine IL-2 (muIL-2), muIL-12p70, muIL-23, and human IL-2 (huIL-2). For the muIL-15 Receptor-oc fused to an IL-15 single chain (muIL- 15Ra-IL- 15sc), an EF-la and CMV promoter were tested. HEK293T STING Null cells (InvivoGen) were seeded in 24-well plates coated with poly-L-lysine at 200,000 cells per well, overnight at 37 °C in a 5% CO2 incubator, to achieve 80% confluency. The following day, 200 ng of each cytokine plasmid DNA was diluted in serum-free media, and added to FuGENE® transfection reagent (Promega), at the proper reagent:DNA ratios, with untransfected wells as negative controls (in duplicates). Cell culture supernatants from each sample were collected at 24 hours post-transfection, and assessed for protein expression by ELISAs specific for each cytokine.

The muIL-2 construct was evaluated in a murine IL-2 ELISA (R&D Systems), according to the manufacturer’s instructions, and an additional version of muIL-2 with codon optimization (muIL-2 CO) also was evaluated. Concentrations of neat supernatant were tested, and yielded an average of 1680 pg/mL of muIL-2 for the muIL-2 construct, and 1812 pg/mL of muIL-2 for the muIL-2 CO construct. These data confirmed the functionality of the constructs, and demonstrated that yield could be improved with codon optimization. The muIL-12p70 construct was evaluated in a murine IL- 12 ELISA (R&D Systems), according to the manufacturer’s protocol. When supernatants were added neat, a mean of 400 pg/mL of secreted muIL-12p70 was measured, although this was outside the linear range. When the supernatants were diluted 5-fold, an average of 105 pg/mL of secreted muIL-12p70 was detected. For the muIL-23 plasmid, detection of protein was achieved using the murine IL-23 ELISA (BioLegend), per kit instructions. With the supernatant added neat, a mean of 966 pg/mL of muIL-23 was detected. For the human IL-2 plasmid, detection of protein was achieved using the human IL-2 ELISA (Invitrogen), per kit instructions. With the supernatant added neat, an average of 1422 pg/mL of huIL-2 was detected.

For the muIL-15Ra-IL-15sc construct, expressed using either the EF-la or CMV promoters, the murine IL- 15 ELISA (eBioscience, Inc.) was used, per kit instructions. When added neat, the muIL-15Ra-IL-15sc plasmid with the EF-la promoter resulted in an average of 131 pg/mL, while the muIL-15Ra-IL-15sc plasmid with the CMV promoter resulted in an average of 289 pg/mL.

These data validate the plasmid expression constructs encoding immunomodulatory cytokines, both mouse and human, in human cells. Further, they indicate that codon optimization, and the use of promoters such as CMV, can enhance protein expression.

Post-Transcriptional Regulatory Elements Enhance Cytokine Expression

In order to determine whether post-transcriptional regulatory elements (PREs), added at the 3’ end of the ORF, enhance expression of immunostimulatory cytokines in human cells, expression of huIL-2, under the control of the EF-la promoter, was tested with or without the addition of a Woodchuck Hepatitis virus post- transcriptional regulatory element (WPRE) in the pATI-1.75 plasmid. HEK293T STING Null cells (InvivoGen) were seeded in 24-well plates coated with poly-L- lysine at 200,000 cells per well, overnight at 37 °C in a 5% CO2 incubator, to achieve 80% confluency. The following day, 200 ng of each cytokine plasmid DNA was diluted in serum-free media and added to FuGENE® transfection reagent (Promega), at the proper reagent:DNA ratios, with untransfected wells as negative controls (in duplicates). Cell culture supernatants from each sample were collected at 24 hours posttransfection and assessed for activity by a human IL-2 ELISA (Invitrogen), according to the manufacturer’s instructions. Supernatants were added neat, or were diluted 5- fold. The results demonstrated that, when supernatant was added neat, compared to the huIL-2 construct without a WPRE, which secreted an average of 1540 pg/mL, the huIL-2 construct with the WPRE secreted 5511 pg/mL, a 3.6-fold increase. In the 5- fold diluted supernatants, the non- WPRE huIL-2 construct secreted 315 pg/mL of huIL-2, while the huIL-2 construct with the WPRE secreted 1441 pg/mL, a 4.6-fold increase. Thus, addition of 3’ post-transcriptional regulatory elements, exemplified by, but not limited to, WPRE, can significantly improve protein expression in human cells.

Promoter Optimization and Post-Transcriptional Regulatory Elements Enhance Cytokine Production in Primary M2 Macrophages

While expression of cytokines, such as muIL-15Ra-IL-15sc, was enhanced in human HEK293T cells by use of the CMV promoter, it was determined whether expression of cytokines could similarly be enhanced in donor-derived primary human M2 macrophages, the predominant macrophage phenotype in T-cell excluded, solid human tumors. Additionally, it was determined whether post-transcriptional regulatory elements, such as WPRE, could enhance expression in these cells.

In order to determine if protein expression could be improved, the promoters EF-la and CMV were tested for controlling expression of muIL-2, and the WPRE post-transcriptional regulatory element was tested for expression of huIL-2. Frozen human PBMCs, isolated from healthy human donors, were thawed in complete medium (RPMI-1640 + IX non-essential amino acids + 5% Human AB serum), and washed by centrifugation for 10 minutes at 800 RPM at room temperature. PBMCs were resuspended in PBS + 2% FBS, and monocytes were negatively isolated using a CD16 depletion kit (StemCell Technologies). The isolated monocytes were cultured for 6 days in RPMI media containing M-CSF and IL-4, to generate M2 macrophages. For this, isolated untouched monocytes were washed by centrifugation in PBS + 2% FBS, and resuspended in complete medium containing 100 ng/mL human M-CSF and 10 ng/mL human IL-4. Isolated monocytes (3e5 per well) were then seeded in a 24- well plate with a final volume of 750 microliters. Two days after the seeding, the cell culture media was entirely aspirated and replaced with fresh complete medium containing 100 ng/mL human M-CSF and 10 ng/mL human IL-4. Two days later (on day 4), 500 pL of complete medium containing 100 ng/mL human M-CSF and 10 ng/mL human IL-4 was added per well, and incubated for 48 hours. On day 6, the cell culture media was entirely aspirated, and replaced with fresh complete medium without cytokines, for transfection with the Viromer® RED mRNA and plasmid transfection reagent (Lipocalyx)

Transfection with Viromer® RED was performed according to the manufacturer’s instructions. Briefly, 500 ng of plasmid DNA, containing the EF-l - muIL-2 construct, the CMV-muIL-2 construct, the EF-l -huIL-2 construct, or the EF- l -huIL-2 + WPRE construct, as well as untransfected control, were diluted in the provided buffer, and mixed with 0.2 pL of Viromer® RED, and incubated at room temperature for 15 minutes to allow the DNA/Viromer® RED complexes to form. The DNA/Viromer® RED complexes were then slowly added to each well of the 24- well plate (in duplicates), and the plate was incubated at 37 °C in a CO2 incubator for 24 hours. Supernatants were harvested at 24 hours, and assayed for cytokines using either a murine IL-2 ELISA (R&D Systems), or a human IL-2 ELISA (Invitrogen), per kit instructions.

The results demonstrated that expression of muIL-2 from neat supernatants harvested from primary human M2 macrophages, transfected with the muIL-2 construct under control of the EF- la promoter, resulted in the secretion of an average of 59.7 pg/mL of muIL-2. The muIL-2 construct with the CMV promoter yielded an average of 275 pg/mL muIL-2, an almost 5-fold increase. For the human IL-2 ELISA, neat supernatants from the cells transfected with the plasmid lacking the WPRE yielded an average of 170 pg/mL huIL-2. The huIL-2 construct containing the WPRE yielded an average of 219 pg/mL huIL-2. These data confirm that promoters such as CMV, and post-transcriptional regulatory elements such as WPRE, can significantly improve cytokine expression in multiple cells types, including primary human M2 macrophages.

Co-Stimulatory Receptor Ligand 4-1BBL Expressed from Human Cells

Co-stimulatory molecules, such as 4-1BBL, when expressed on antigen- presenting cells (APCs), can engage 4-1BB expressed on T-cells to promote optimal T-cell function. 4-1BBL is negatively regulated by its cytoplasmic signaling domain. In the late-phase of 4-1BBL ligation of macrophages to T-cells, reverse signaling of the 4-1BBL cytoplasmic domain induces surface translocation of 4-1BBL to bind and form a signaling complex with TLR4. This induces high levels of TNF-a, comparable to LPS activation of TLR4, that leads to immunosuppression of the adaptive immune response (see, e.g., Mae/ al. (2013) Set. Signal. 6(295):ra87).

In this example, the sequence encoding murine 4-1BBL was cloned into the pATI-1.75 vector. In order to maximally engage T-cells, the reverse signaling of the 4-1BBL cytoplasmic domain was eliminated by deleting the cytoplasmic domain (corresponding to amino acid residues 1-82 of SEQ ID NO:344), generating mu4- IBBLAcyt. To determine whether mu4-lBBLAcyt could be functionally expressed on the surface of human cells, HEK-293T cells were utilized. HEK293T STING Null cells (InvivoGen) were seeded in 24-well plates coated with poly-L-lysine at 200,000 cells per well, overnight at 37 °C in a CO2 incubator, to achieve 80% confluency. The following day, 200 ng of plasmid DNA, encoding mu4-lBBLAcyt, was diluted in serum-free media and added to FuGENE® transfection reagent (Promega), at the proper reagent:DNA ratio, with untransfected wells as a negative control (in duplicates). After 48 hours, the cells were washed twice with PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes. The cells were then resuspended in PBS + 2% FBS, and stained with a PE-conjugated murine anti-4-lBBL antibody (clone TKS-1, BioLegend) and DAPI (dead/live stain). After 30 minutes, the cells were washed twice with PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes, and resuspended in PBS + 2% FBS. Flow cytometry data were acquired using the ACEA NovoCyte® flow cytometer (ACEA Biosciences, Inc.), and analyzed using the FlowJo™ software (Tree Star, Inc ).

As a percentage of live cells, the untransfected control cells showed a percent positive staining for murine 4-1BBL of 14.6. In comparison, 93.4% of the cells that were transfected with the plasmid encoding mu4-lBBLAcyt, were positive for surface expression of 4-1BBL. These data demonstrate that the pATLl .75 plasmid can effectively be used to express 4-1BBL at high levels on the surface of human cells. Soluble TGFp Receptor II Expressed from Human Cells

Soluble mouse TGFp receptor II variants were designed by removing the cytoplasmic and transmembrane portions of the full TGF receptor II. Additionally, either a FLAG or Fc tag was added for detection. These variants were cloned into the pATI-1.75 vector under the control of a CMV promoter and a 3’ WPRE. The sequences were confirmed by Sanger sequencing. 1.5 x 10 ⁶ HEK293T cells were plated one day prior on 6-well plates coated with poly-L-lysine, to achieve 80% confluency. On the day of transfection, 3 jag of DNA was diluted in serum-free media and added to FuGENE® transfection reagent (Promega) at the proper reagent:DNA ratios. Cell culture supernatants from each sample were collected after 48 hours of incubation. Some supernatant was concentrated in a 10 EDa spin column (Millipore). Direct ELISAs with mouse TGF-pl (R&D systems) were performed on the supernatant of transfected HEK293T cells. ELISA data, with absorbance at 450 nm, is provided in the table below.

The functionality of these constructs was tested in a T-cell assay. Mouse T- cells were harvested from the spleen using a magnetic isolation kit (StemCell Technologies). T-cells were incubated with anti-mouse CD3s antibody, with or without the soluble receptor, at various concentrations of mouse TGF-beta. T-cell activation was quantified using the mouse TH1 CBA kit (BioLegend), and flow cytometry labeling of CD4, CD8, 4-1BB, and CD69.

These data demonstrate the ability to express heterologous molecules, such as extracellular receptors fused to an Fc domain, from the plasmid engineered for delivery by the immunostimulatory bacteria to eukaryotic, such as human, cells. Expression of a CD3xCD19 Bispecific T-cell Engager from Human Cells

A CD3xCD19 bispecific T-cell engager (BiTE®), containing a FLAG tag and a His tag, was cloned into the p ATI- 1.75 vector under the control of a CMV promoter and with a 3’ WPRE. The sequences were confirmed by Sanger sequencing. 1.5 x 10 ⁶ HEK293T cells were plated one day prior on 6-well plates coated with poly-L-lysine, to achieve 80% confluency. On the day of transfection, 3 pg of DNA was diluted in serum-free media and added to FuGENE® transfection reagent (Promega), at the proper reagent:DNA ratios. Cell culture supernatants from each sample were collected after 48 hours of incubation. Some supernatant was concentrated in a 10 EDa spin column (Millipore).

The functionality of this construct was tested by binding of the CD3xCD19 BiTE® to Raji and JurEat-Lucia™ NF AT cells (InvivoGen). The BiTE® was detected using an anti-FLAG-APC (BioLegend), using flow cytometry. One well of 50,000 cells was run for each condition. The mean fluorescence intensity (MFI) of the APC positive events, and the number of cells gated as APC positive, are provided in the table below.

Additionally, this construct was tested in a co-culture of Raji and Jurkat- Lucia™ NF AT cells. Cells were incubated with or without the CD3xCD19 BiTE®, and the luminescence (corresponding to Lucia™ luciferase activity resulting from NF AT activation) was detected at 6 hours and 24 hours post-addition of the BiTE®. The luminescence readings of this assay are provided in the table below. These data demonstrate the ability to express heterologous molecules, such as scFvs, alternative antibody constructs, and bispecific T-cell engagers, in eukaryotic, such as human, cells, from the engineered plasmid that can be delivered by the immunostimulatory bacteria herein. EXAMPLE 10

Immunostimulatory Bacterial Strains Efficiently Deliver Plasmids and Express Cytokines in Human Cells

Flagella-Deleted Strains Containing Plasmids Encoding Murine IL-2 Induce Functional IL-2 Protein Expression Following Infection in Human Monocytes

As described above, the flagellin genes fljB and fliC were deleted from the YS1646 strain of S. typhimurium with the asd gene deleted, generating the strain YS 1646Aa.st//AFLG. This strain was electroporated with a plasmid containing an expression cassette with the EF-la promoter and the murine cytokine IL-2 (muIL-2). In addition, the YS1646Am<7/AFLG strain was electroporated with an expression plasmid encoding murine IL-153, as a control for a non-cognate cytokine. Additional constructs were created using the CMV promoter.

To determine whether these strains containing expression plasmids can infect human monocytes and induce the production of murine IL-2, THP-1 human monocytic cells were plated at 50,000 cells/well in RPMI 1640 (Gibco™) + 10% Nu- Serum™ (Coming®), one day prior to infection. The cells were infected with the various strains at an MOI of 50 for one hour in RPMI, then washed 3 times with PBS, and resuspended in RPMI + 100 pg/mL gentamicin (Sigma). Supernatants were collected 48 hours later from a 96-well plate, and assessed for the concentration of murine IL-2 by ELISA (R&D Systems).

The concentration of muIL-2 detected in the YS I646Ar/.sc//AFLG-IL- l 53 control wells was very low (6.52 pg/mL), as expected, and reflective of some crossreactivity to the endogenous human IL-2 receptor. In contrast, the YS 1646Ac/.st/AFLG-muIL-2 strain induced an average of 35.1 pg/mL of muIL-2. These data demonstrate the feasibility of expressing and secreting functional heterologous proteins, such as IL-2, from the 5. typhimurium immunomodulatory platform strains, in human monocytes.

Flagella-Deleted and p -Deleted Strains, Containing Plasmids Encoding Murine IL-2, Demonstrate Enhanced IL-2 Expression, Compared to Transfected muIL-2 DNA, in Primary Human M2 Macrophages

The relative efficiencies of transfection (i.e., direct transfer of plasmid DNA) vs. bactofection (i.e., transfer of plasmid DNA by the immunostimulatory bacterial strains herein), in primary human M2 macrophages, for expression of muIL-2, were compared. Frozen human PBMCs, isolated from healthy human donors, were thawed in complete medium (RPMI-1640 + IX non-essential amino acids + 5% Human AB serum), and washed by centrifugation for 10 minutes at 800 RPM at room temperature. PBMCs were resuspended in PBS + 2% FBS, and monocytes were negatively isolated using a CD 16 depletion kit (StemCell Technologies). Isolated untouched monocytes were then washed by centrifugation in PBS + 2% FBS, and resuspended in complete medium containing 100 ng/mL human M-CSF and 10 ng/mL human IL-4. Isolated monocytes (3e5 per well) were then seeded in a 24-well plate, with a final volume of 750 microliters. Two days after the seeding, the cell culture media was entirely aspirated and replaced with fresh complete medium containing 100 ng/mL human M-CSF and 10 ng/mL human IL-4. Two days later (on day 4), 500 pL of complete medium containing 100 ng/mL human M-CSF and 10 ng/mL human IL-4 was added per well, and incubated for 48 hours. On day 6, the cell culture media was entirely aspirated, and replaced with fresh complete medium without cytokines, for transfection with Viromer® RED mRNA and plasmid transfection reagent (Lipocalyx).

Transfection with Viromer® RED was performed according to the kit instructions. Briefly, 500 ng of plasmid DNA from the constructs encoding muIL-2 under control of the EF-la promoter (EF-la-muIL-2), or under control of the CMV promoter (CMV-muIL-2), or untransfected control, were diluted in the provided buffer, mixed with 0.2 pL of Viromer® RED transfection reagent, and incubated at room temperature for 15 minutes to allow the DNA/Viromer® RED complexes to form. The DNA/Viromer® RED complexes were then slowly added to each well of the 24-well plate containing the monocytes (in duplicates), and the plate was incubated at 37 °C in a CO2 incubator for 24 hours. Additional wells of cells were infected at an MOI of 450 with the YS1646Ar/.s /AFLG/A/«/p/-’ strain containing the EF-la-muIL-2 construct, or the YS1646 tz t//AFLG/A ?tzgP strain containing the CMV-muIL-2 construct, for one hour in RPMI, then washed 3 times with PBS, and resuspended in RPMI + 100 pg/mL gentamicin (Sigma).

After 24 hours, the cells were lysed with 350 pL Buffer RLT with [3- mercaptoethanol (p-ME) (Qiagen), and RNA extraction was performed using the Qiagen RNeasy® Mini Kit with the following modifications. A genomic DNA elimination step, using an RNase-Free DNase kit (Qiagen) was included in the kit to remove genomic DNA from the total RNA. Total RNA concentration was measured using a NanoDrop™ One ^c UV-Vis Spectrophotometer (Thermo Fisher Scientific). The purity of each sample also was assessed from the A260/A230 absorption ratio. RNA was stored at -80 °C without freeze-thawing until reverse-transcription was performed. cDNA synthesis was performed using 0.4-1 pg of template RNA using a Cl 000 Touch Thermal Cycler (Bio-Rad) and SuperScript™ VILO™ Master Mix (Invitrogen) in a 30 pL reaction, according to the manufacturer’s instructions. qPCR (quantitative polymerase chain reaction) was performed with a CFX96™ Real-Time PCR Detection System (Bio-Rad). SYBR® primers for murine IL-2 (Assay ID: qMmuCED0060978) were purchased from Bio-Rad. The qPCR reaction (20 pL) was conducted per protocol, using the iTaq™ Universal SYBR® Green Supermix (Bio-Rad). The standard thermocycling program on the Bio-Rad CFX96™ Real-Time System consisted of a 95 °C denaturation for 30 seconds, followed by 40 cycles of 95 °C for 5 seconds and 60 °C for 30 seconds. Reactions with template-free control were included for each set of primers on each plate. All samples were run in duplicate, and the mean C _q values were calculated. Quantification of the target mRNA was normalized using Gapdh (glyceraldehyde-3-phosphate dehydrogenase) reference mRNA (Bio-Rad, Assay ID: qMmuCED0027497). AC _q was calculated as the difference between target (muIL-2) and reference (GAPDH) gene. AAC _q was obtained by normalizing the AC _q values of the treatments, to the AC _q values of the non-treatment controls. Fold increase was calculated as 2 ^A -AAC _q . The fold increases relative to untransfected/uninfected control are shown in the table below.

The results show that, with either transfection or bactofection, the CMV promoter demonstrated superior expression of muIL-2, compared to the EF-lot promoter, in primary human M2 macrophages. While transfection using the most efficient reagent currently available gave the highest levels of muIL-2 expression, bactofection also elicited high expression levels of muIL-2, demonstrating the high efficacy of heterologous gene transfer with the bacterial platforms provided herein. EXAMPLE 11

Bacterial Strains Efficiently Deliver Immunomodulatory Plasmids In Vivo and Demonstrate Potent Anti-tumor Activity

Flagella-Deleted Strains Containing Plasmids Encoding Murine IL-2 Induce Potent Tumor Inhibition without Toxicity in a Mouse Model of Colorectal Carcinoma

In order to demonstrate that S. typhimurium strains containing the muIL-2 expression plasmids can induce anti-tumor efficacy without additional toxicity, the YS1646Aa«//AFLG strain, containing the muIL-2 plasmid, was compared to PBS control, for safety and efficacy in the subcutaneous flank MC38 colorectal adenocarcinoma model. For this study, 6-8 week-old female C57BL/6 mice (5 mice per group) were inoculated SC in the right flank with MC38 cells (5xl0 ⁵ cells in 100 pL PBS). Mice bearing established flank tumors were IV injected on day 11 with 5xlO ⁵ CFUs of the YS1646Aa«//AFLG-muIL-2 strain, or with PBS vehicle control. Tumor measurements and body weights were recorded twice weekly.

The results revealed that the YS I646Ar/.s /AFLG-muIL-2 strain demonstrated significant tumor growth inhibition (TGI) compared to PBS (76.7% TGI, P = 0.005, day 21), with tumors being well controlled out to day 40 post-implantation, when the PBS mice were euthanized. The therapy was well tolerated, even without further strain attenuation, and the weight loss early on was transient and resulted in only a 3.4% reduction in body weight, compared to PBS control at day 40. Thus, the immunostimulatory strain expressing muIL-2 potently inhibits tumor growth inhibition, in a safe and non-toxic manner, in a model of colorectal carcinoma. Flagella-Deleted Strains Containing Plasmids Encoding Murine IL-2 Induce Tumor-Specific Production of IL-2 In Vivo

The level of tumor muIL-2 expression, relative to spleen, was determined in order to confirm the tumor-specific nature of delivery. 6-8 week-old female C57BL/6 mice (5 mice per group) were inoculated SC in the right flank with MC38 colorectal adenocarcinoma cells (5xlO ⁵ cells in 100 pL PBS). Mice bearing established flank tumors were IV injected on day 10 with 5xl0 ⁵ CFUs of the YS1646Aast//AFLG- muIL-2 strain, or with PBS vehicle control. On day 31 post tumor implantation, tumors and spleens were excised and processed for tumor extracts using the GentleMACS™ Octo Dissociator and the M tubes (Miltenyi Biotec) molecule setting in 2 mL of PBS. The homogenates were spun down at 1300 RPM for 10 minutes, and the supernatant was collected and assayed using the muIL-2 CBA kit (BD Biosciences), according to the manufacturer’s instructions Results were quantified as pg/mL of muIL-2, and standardized to per gram of tissue.

The PBS control tumors exhibited background levels of muIL-2 in the tumor, with a mean of 134 pg/mL per gram of tumor tissue. The YS I 646Aast//AFLG-muIL-2 treated tumors yielded a much higher mean of 389.9 pg/mL of muIL-2 per gram of tumor tissue, demonstrating the ability to detect elevated muIL-2 levels due to plasmid delivery in the tumor-resident myeloid cells. The level of muIL-2 in the spleen, from mice injected with the YS l 646A«.sz//AFLG-muIL-2 strain, was an average of 6.6 pg/mL per gram of tissue, which was lower than in the PBS controls. This specificity for the tumor enables delivery of immunomodulatory levels of IL-2, in a much safer manner than conventional cytokine therapies, which are not tumor- targeted.

Attenuated Bacterial Strains Containing Plasmids Encoding Murine Costimulatory Receptor Ligand 4-1BBL Demonstrate Curative Effects In Vivo

In order to determine whether tumor-specific delivery of a co-stimulatory molecule, such as 4-1BBL, enhances anti -turn or efficacy, a bacterial strain containing a plasmid encoding 4-lBBL(Acyt) (described above), under control of the CMV promoter and containing a 3’ WPRE, was assessed in the MC38 murine model of colorectal adenocarcinoma. For this study, 6-8 week-old female C57BL/6 mice (5 mice per group) were inoculated SC in the right flank with MC38 colorectal adenocarcinoma cells (5xlO ⁵ cells in 100 pL PBS). Mice bearing established flank tumors were IV injected on day 10 with IxlO ⁷ CFUs of strain YS I646A«.szZ ^/ AFLG/A/?« / ^> /Ar/».sA containing the CMV-4-lBBL(Acyt)-WPRE plasmid, or with PBS vehicle control.

The therapy was very well tolerated, with only an initial weight loss of 2.2% that had fully recovered 3 days later. Compared to PBS, the 4-lBBL(Acyt) therapy was highly effective and curative (90.7% TGI, 60% complete response (CR), day 30). These data demonstrate the potency and safety of delivering immunostimulatory bacteria containing plasmids encoding co-stimulatory molecules in a tumor-specific manner. EXAMPLE 12

Identification of Gain-of-Function Mutations in Genes That Promote Constitutive Type I Interferon Production

Instances of subjects presenting with severe auto-inflammatory conditions and vasculopathies of unknown etiology occur, and, often derive from mutations. The cause for these conditions has, and can be, identified. Steps to identify a mutational basis for such a pathology are as follows. In step one, intact genomic DNA is obtained from patients experiencing symptoms, and from healthy individuals. Genome sequence is performed, such as whole exome sequencing, and then introns and exons are analyzed. Analysis of genes, and identification of mutations in products in the pathways associated with the expression of type I interferon (IFN), is performed. From this analysis, mutations are discovered in genes known to lead to constitutive functional activation of the encoded proteins, and subsequent persistent expression of type I IFN.

After identification of mutations, cDNAs encoding the full-length gene, with and without the identified mutation(s), are transfected into a reporter cell line that measures expression of type I IFN. For example, a reporter cell line can be generated where the expression of luciferase is placed under control of the promoter for IFN-p. A gain-of-function (GOF) mutant that is constitutively active will promote the expression of IFN-P, whereas the unstimulated wild-type (WT) protein will not. In the case of known STING SAVI (STING-associated vasculopathy with onset in infancy) mutants, the WT STING stimulation of IFN-P requires the addition of increasing exogenous levels of cGAMP to directly activate WT STING. Constitutively active mutations stimulate the expression of IFN-P in a cGAMP-independent manner. Exemplary gain-of-function mutations in each of STING, RIG-I, MDA5, IRF3, and IRF7 are set forth below in Example 15, and discussed elsewhere herein. Other such genes, in which gain-of-function mutations can be identified in subjects, or produced by in vitro mutation and screening, include, but are not limited to, TRIM56, RIP1, Sec5, TRAF3, TRAF2, TRAF6, STAT1, LGP2, DDX3, DHX9, DDX1, DDX9, DDX21, DHX15, DHX33, DHX36, DDX60, and SNRNP200.

Expression of Functional Constitutive Type I IFN Mutants in Human Cells

Human STING (allele R232) and IRF3 gain-of-function (GOF) mutants (see, table below) were cloned into the pATI-1.75 vector, and the sequences were confirmed by PCR. To determine whether the STING and IRF3 GOF expression plasmids could induce functional type I IFN in human cells, the plasmids were assessed using HEK293T STING Null Reporter cells (InvivoGen), which do not contain endogenous STING. These cells express secreted embryonic alkaline phosphatase (SEAP), placed under the control of the endogenous IFN-stimulated response element (ISRE) promoter, where the coding sequence of ISRE has been replaced by the SEAP ORF using knock-in technology. Type I interferon activity can be assessed by monitoring type I IFN-stimulated SEAP production in the cell culture supernatants.

To test the relative production of type I IFN by each of the GOF mutants, IxlO ⁵ 293T-Dual™ Null cells (InvivoGen) were plated one day prior on plates coated with poly-L-lysine, to achieve 80% confluency, in a 24-well plate. On the day of transfection, 200 ng of plasmids encoding a panel of STING and IRF3 GOF mutants, including a STING wild-type (WT) and IRF3 WT control, and a negative control STING mutant that has been reported in the literature to be non-functional in human cells (STING VI 55R negative control (NC)), were diluted in serum-free media and added to FuGENE® transfection reagent (Promega) at the proper reagent:DNA ratios. Cell culture supernatants from each sample were collected after overnight incubation, and 10 pL of the cell culture supernatants was added to 50 pL QUANTI-Blue™ reagent (InvivoGen), which is used for measuring SEAP. Type I interferon activation was determined by measuring ISRE-induced SEAP activity on a SpectraMax® M3 Spectrophotometer (Molecular Devices), at an absorbance wavelength of 650 nm.

As shown in the table below, all GOF mutants were able to induce type I IFN activity in a STING ligand-independent manner in human cells, compared to the wildtype and negative controls, which did not induce type I IFN activity. The highest levels of type I IFN induction were observed with the human STING R284G variant, and the human IRF3 S396D phosphomimetic variant. These data support the ability of the plasmids encoding GOF mutants to produce functional, constitutively active STING and constitutively active phosphomimetic IRF3, that can induce type I IFN in a cGAMP -independent manner.

Infection of Flagella-Deleted Strains, Containing Plasmids Encoding Constitutively Active Type I IFN Mutants, Converts Human M2 Macrophages to Type I IFN-Producing Ml Macrophages

It was determined if primary human M2 macrophages, infected with flagella- deleted strains containing plasmids encoding constitutively active type I IFN GOF variants, could be converted to producers of type I IFN and downstream chemokines, such as CXCL10 (also known as IP- 10).

Frozen human PBMCs, isolated from healthy human donors, were thawed in complete medium (RPMI-1640 + IX non-essential amino acids + 5% Human AB serum), and washed by centrifugation for 10 minutes at 800 RPM at room temperature. PBMCs were resuspended in PBS + 2% FBS, and monocytes were negatively isolated using a CD 16 depletion kit (StemCell Technologies). To generate primary human M2 macrophages, isolated untouched monocytes were washed by centrifugation in PBS + 2% FBS, and resuspended in complete medium containing 100 ng/mL human M-CSF and 10 ng/mL human IL-4. Isolated monocytes (3e5 per well) were then seeded in a 24-well plate with a final volume of 750 microliters. Two days after the seeding, the cell culture media was entirely aspirated and replaced with fresh complete medium containing 100 ng/mL human M-CSF and 10 ng/mL human IL-4. Two days later (on day 4), 500 pL of complete medium containing 100 ng/mL human M-CSF and 10 ng/mL human IL-4 was added per well, and incubated for 48 hours. On day 6, the cell culture media was entirely aspirated and replaced with fresh complete medium without cytokines. Duplicate wells were infected at an MOI of 450, for one hour in RPMI, with the following strains: YS 1646A< .sz//AFLG containing a plasmid encoding wild-type (WT) human (hu) STING; YS1646Aa%7/AFLG containing a plasmid encoding the huSTING R284G variant; YS 1646A .st//AFLG containing a plasmid encoding WT huIRF3; YS1646Aasz//AFLG containing a plasmid encoding the huIRF3 S396D variant; or a strain containing a plasmid control. The cells were then washed 3 times with PBS, and resuspended in RPMI + 100 pg/mL gentamicin (Sigma). As a control, the STING agonist 3’5’ RpRp c-di-AMP (InvivoGen), an analog of the clinical compound ADU-S100, was added to the cells at a concentration of 10 pg/mL.

After 24 hours, the cells were lysed with 350 pL Buffer RLT with P-ME (Qiagen), and RNA extraction was performed using the Qiagen RNeasy® Mini Kit with the following modification. A genomic DNA elimination step, using an RNase- Free DNase kit (Qiagen), was included to remove genomic DNA from the total RNA. Total RNA concentration was measured using a NanoDrop™ One ^c UV-Vis Spectrophotometer (Thermo Scientific). The purity of each sample also was assessed from the A260/A230 absorption ratio. RNA was stored at -80 °C without freeze-thawing until reverse-transcription was performed. Synthesis of cDNA was performed from 0.4-1 pg of template RNA using a C1000 Touch Thermal Cycler (Bio-Rad) and Super Seri pt™ VILO™ Master Mix (Invitrogen) in a 30 pL reaction, according to the manufacturer’s instructions. qPCR was performed with a CFX96™ Real-Time System (Bio-Rad). SYBR® primers for huCXCLIO (qHsaCED0046619), huIRF3 (qHsaCID0013122), huSTING (qHsaCIDOO 10565), and huIFN i (qHsaCED0046851) were purchased from BioRad. The qPCR reaction (20 pL) was conducted per protocol, using the iTaq™ Universal SYBR® Green Supermix (Bio-Rad). The standard thermocycling program on the BioRad CFX96™ Real-Time System consisted of a 95 °C denaturation for 30 seconds, followed by 40 cycles of 95 °C for 5 seconds and 60 °C for 30 seconds. Reactions with template free control were included for each set of primers on each plate. All samples were run in duplicate, and the mean C _q values were calculated. Quantification of the target mRNA was normalized using Gapdh reference mRNA (Bio-Rad, qMmuCED0027497). AC _q was calculated as the difference between the target and reference gene. AAC _q was obtained by normalizing the AC _q values of the treatments to the AC _q values of the non-treatment control. Fold increase was calculated as 2 ^A -AAC _q . The values are shown in the table below, as the average of the duplicate wells.

As shown in the table below, compared to the infection of the plasmid control, strains of YS1646Aas<7/AFLG, containing plasmids encoding WT huSTING and huSTING R284G, induced high levels of STING expression, which were significantly higher compared to the plasmid control, or the small molecule STING agonist. Similarly, the strains containing plasmids encoding WT huIRF3 and huIRF3-S396D induced high levels of IRF3 expression, which were significantly higher than the plasmid control, or the small molecule STING agonist. The bacterial strain containing a plasmid encoding the huSTING R284G variant induced much higher expression of IFNP and CXCL10, as compared to the strain containing a plasmid encoding WT huSTING. This demonstrates the ability of the strain, containing a plasmid encoding a constitutively active STING GOF variant, to convert a human primary, immunosuppressive M2 macrophage, into an Ml, type I IFN producing, cell. While the strains containing plasmids encoding WT huIRF3 and huIRF3-S396D both induced more, or similar levels of IFNP, they induced less CXCL10 than the huSTING-R284G variant.

ND = No Data

These data demonstrate the expression of constitutively active GOF type I IFN variants in human primary M2 macrophages, and converting these cells to Ml -like, type I IFN producing, cells.

EXAMPLE 13

Immunostimulatory Bacteria, Containing Plasmids Encoding Constitutive Type I IFN Variants, Demonstrate Potent Anti-tumor Immunity in a Murine Model of

Colorectal Cancer

Human GOF STING Mutants Show Anti-Tumor Activity in Mouse Models

To demonstrate that immunostimulatory bacterial strains, containing expression plasmids encoding constitutively active STING variants, induce anti-tumor efficacy, strain YS I646A .st//AFLG (with a knockout of both flagellin genes fljB and fliC) was electroporated with a plasmid containing an expression cassette for human STING with the allele R232 and the GOF mutation V155M (huSTING V155M), behind the human elongation factor-1 alpha (EF-la) promoter, and was compared to strain YS1646 alone, and to a PBS vehicle control. The gene encoding huSTING V155M was generated using DNA synthesis and cloned into the pATI-1.75 vector. In order to evaluate whether a constitutively active human STING variant could demonstrate anti-tumor activity in mice, 6-8 week-old female C57BL/6 mice (5 mice per group) were inoculated SC in the right flank with MC38 colorectal adenocarcinoma cells (5xlO ⁵ cells in 100 pL PBS). Mice bearing established flank tumors were IV injected on day 8 with 5xl0 ⁵ CFUs of strain YS 1646A< .sc//AFLG- huSTING V155M, with strain YS1646, or with PBS control.

The results showed that the YS1646 parental strain was only mildly effective as an anti -turn or therapy, and was not curative (35% TGI, p = NS (not significant), day 28), in line with previously published data. The more attenuated strain, containing a plasmid encoding constitutively active human STING, YS I646Aa.st//AFLG- huSTING V155M, however, elicited significant tumor control (60% TGI,/? < 0.05, day 28) compared to PBS, and had a cure rate of 20%. Thus, an immunostimulatory bacterial strain that delivers a constitutively active STING variant potently inhibits tumor growth, and demonstrates curative effects in a model of colorectal adenocarcinoma.

Murine Phosphomimetic IRF3 Shows Curative Effects In Vivo

The murine version of the phosphomimetic human IRF3 variant was designed, designated muIRF3-S388D, and evaluated in a murine model of colorectal adenocarcinoma. Strain YS I646Ac/.st//AFLG was electroporated with a plasmid containing an expression cassette for murine IRF3 with the GOF mutation S388D (muIRF3-S388D), behind the human elongation factor-1 alpha (EF-la) promoter, and was compared to PBS vehicle control. The gene encoding muIRF3-S388D was generated using DNA synthesis and cloned into the pATI-1.75 vector. 6-8 week-old female C57BL/6 mice (5 mice per group) were inoculated SC in the right flank with MC38 colorectal adenocarcinoma cells (5xI0 ⁵ cells in 100 pL PBS). Mice bearing established flank tumors were IV injected on day 10 with 5 xlO ⁵ CFUs of strain YS1646Aa«//AFLG-EF-la-muIRF3-S388D, and compared to PBS vehicle control.

The therapy was very well tolerated, with an initial weight loss nadir of only 0.3%. Compared to PBS, the bacterial strain containing the plasmid encoding the muIRF3-S388D GOF mutant was highly effective and curative (81.8% TGI, 60% CR, day 42). These data demonstrate the potency and safety of delivering constitutively active type I IFN inducing variants in a tumor-specific manner. Murine STING GOF Variants Show Potent and Curative Anti-Tumor Activity

A panel of murine orthologs of the human STING variants, discovered in human patients, was designed. These orthologs differ by one codon from the human variants, and were cloned into the pATI-1.75 vector under the control of an EF-la promoter, to yield the following set of mutants: muSTING N153S, muSTING V154M, muSTING R280Q, muSTING V146L, muSTING R283G, and muSTING C205Y, among others. The STING variants were evaluated in the MC38 model of murine adenocarcinoma for anti -turn or efficacy. For the studies, 6-8 week-old female C57BL/6 mice (5 mice per group) were inoculated SC in the right flank with MC38 colorectal adenocarcinoma cells (5xl0 ⁵ cells in 100 pL PBS). Mice bearing established flank tumors were IV injected on day 10 with 5 xlO ⁵ CFUs of strain YS 1646Aa.se/AFLG, containing a plasmid with EF-la driving the expression of muSTING N153 S, muSTING VI 54M, muSTING R280Q, muSTING V146L, or muSTING R283G, or a scrambled shRNA plasmid control, and compared to PBS vehicle control.

In this experiment, strain YS I646Aa.w//AFLG EF- l a-shSCR (scrambled plasmid control) demonstrated anti -tumor efficacy as compared to PBS control (73% TGI, day 26), which was much more potent than the YS1646 parental strain has shown historically. This can be due to inherently immunostimulatory elements on the plasmid itself, such as CpGs and RNAi stimulatory elements. However, this therapy was the least well tolerated of the group, demonstrating a weight loss nadir of 9.9% that only resolved at the very end of the study. In contrast, the constitutively active murine STING mutants resulted in a lower weight loss that was transient and that resolved within days. The relative anti-tumor efficacy of these variants revealed differences in activity, with only two variants demonstrating curative effects and enhanced efficacy over the plasmid control, muSTING N153 S, and muSTING R283G. In a follow-up study, the murine STING C205Y variant was tested along with the R283G and N153S variants, to compare their anti-tumor efficacy. 6-8 week-old female C57BL/6 mice (5 mice per group) were inoculated SC in the right flank with MC38 colorectal adenocarcinoma cells (5xl0 ⁵ cells in 100 pL PBS). Mice bearing established flank tumors were IV injected on day 9 with 5 xlO ⁵ CFUs of strain YS1646Aayr//AFLG containing a plasmid with EF-la driving the expression of muSTING N153S, muSTING R283G, or muSTING C205Y, and compared to PBS vehicle control. As before, the STING variants were well tolerated, and only a transient dip in weight loss was observed that resolved quickly. This is likely due to on-target therapy, as it is also observed with the small molecule STING agonists. The efficacy of the two constitutively active murine STING variants, muSTING N153 S and muSTING R283G, was nearly identical to the previous study, although the weight loss was much less, for reasons unclear. The muSTING C205Y variant also was highly effective, although not curative.

The STING-cured mice from these studies were re-challenged at day 40 postinitial tumor implantation on the opposite flank, SC with MC38 colorectal adenocarcinoma cells (5xlO ⁵ cells in 100 pL PBS). Compared to naive mice (N=5), in which all tumors grew out, all of the STING-cured mice rejected the tumors, demonstrating the engagement of adaptive immunity.

These data validate the safety and potency of the murine versions of the human constitutively active STING variants in a murine model of colorectal carcinoma, and reveal a small subset of variants that have enhanced potency compared to the other STING variants. These highly active variants also elicit protective immunity, demonstrating the potency of tumor-specific production of type I interferon.

Murine STING GOF Variants Demonstrate Significant Tumor Remodeling Following IV Dosing

It was next determined whether the bacterial strains containing plasmids encoding constitutively active STING variants demonstrate differences in their ability to remodel the tumor microenvironment (TME) following IV dosing. To test this, 6-8 week-old female C57BL/6 mice (5 mice per group) were inoculated SC in the right flank with MC38 colorectal adenocarcinoma cells (5xl0 ⁵ cells in 100 pL PBS). Mice bearing established flank tumors were IV injected on day 8 with 5 xlO ⁵ CFUs of strain YS1646A< .st//AFLG, containing a plasmid with EF-la driving the expression of muSTING N153 S, muSTING VI 54M, muSTING R280Q, muSTING V146L, muSTING R283G, or plasmid control, and compared to PBS vehicle control.

At day 28 post tumor implantation, tumors were excised for analysis. Tumors were cut into 2-3 mm pieces into gentleMACS™ C tubes (Miltenyi Biotec) filled with 2.5 mL enzyme mix (RPMI-1640 + 10% FBS with 1 mg/mL Collagenase IV and 20 pg/mL DNase I). The tumor pieces were dissociated using OctoMACS™ (Miltenyi Biotec) specific dissociation program (mouse implanted tumors), and the whole cell preparation was incubated with agitation for 45 minutes at 37 °C. After 45 minutes of incubation, a second round of dissociation was performed using the OctoMACS™ (mouse implanted tumor) program, and the resulting single cell suspensions were filtered through a 70 pM nylon mesh into a 50 mL tube. The nylon mesh was washed once with 5 mL of RPMI-1640 + 10% FBS, and the cells were filtered a second time using a new 70 pM nylon mesh into a new 50 mL tube. The nylon mesh was washed with 5 mL of RPMI-1640 with 10% FBS, and the filtered cells were then centrifuged at 1000 RPM for 7 minutes. The resulting dissociated cells were resuspended in PBS and kept on ice before the staining process.

The percentage of live tumor-infiltrating leukocytes (TILs), including CD4 ⁺ Tregs, CD4 ⁺ Thl cells, CD8 ⁺ T cells, neutrophils, monocytes, dendritic cells (DCs), Ml macrophages and M2 macrophages, following the administration of strain YS1646 a%//AFLG, containing plasmids encoding the various GOF muSTING mutants, was determined by flow cytometry. For the flow-cytometry staining, 100 pL of the single cell suspensions were seeded in wells of a V-bottom 96-well plate. PBS containing a dead/live stain (Zombie Aqua™, BioLegend) and Fc Blocking reagents (BD Biosciences) were added at 100 uL per well, and incubated on ice for 30 minutes in the dark. After 30 minutes, the cells were washed twice with PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes. The cells were then resuspended in PBS + 2% FBS, containing fluorochrome-conjugated antibodies (CD4 FITC clone RM4-5; CD8a BV421 clone 53-6.7; F4/80 APC clone BM8; CD1 lb PE-Cy7 clone MI/70; CD45 BV570 clone 30-F11; CD3 PE clone 145-2C11; Ly6C BV785 clone HK1.4; I- A/I-E APC-Cy7 clone M5/114.15.2; Ly6G BV605 clone 1A8; and CD24 PercP- Cy5.5 clone MI/69; all from BioLegend), and incubated on ice for 30 minutes in the dark. After 30 minutes, the cells were washed twice with PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes and resuspended in flow cytometry fixation buffer (Thermo Fisher Scientific). Flow cytometry data were acquired using the ACEA NovoCyte® flow cytometer (ACEA Biosciences, Inc.), and analyzed using the FlowJo™ software (Tree Star, Inc ).

As shown in the tables below, the YS1646Aa%Z/AFLG strain with the EF-la plasmid control demonstrated predominantly high neutrophil infiltration, despite some CD8 ⁺ T-cell recruitment, likely due to immunostimulatory elements on the plasmid. In contrast, the different muSTING variants had unique tumor-infiltrating immune cell signatures, with some, such as muSTING V146L and muSTING R283G resulting in fewer immunosuppressive neutrophils than the PBS control. The most favorable immune profiles were observed in the tumors from mice that were administered the muSTING R283G and muSTING N153 S mutants, with high numbers of CD4 ⁺ Thl cells and CD8 ⁺ T-cells, and low numbers of neutrophils, which indicates highly favorable conditions for generating an adaptive immune response. In addition, the muSTING R283G and muSTING N153 S mutants had significantly higher pl 5e tumor-antigen-specific CD8 ⁺ T-cells in the tumor, as compared to PBS. These trends were also recapitulated in the total cell counts, as shown below. Thus, delivery of constitutively active STING variants to the tumor-resident myeloid cells leads to a complete remodeling of the immunosuppressive tumor microenvironment, towards an adaptive anti-tumor phenotype, and away from a bacterial phenotype, which is characterized by the promotion of innate immunity and the suppression of adaptive immunity.

% of Live Tumor-Infiltrating Leukocytes (TILs)

Total Cell Counts EXAMPLE 14

Expression of Multiple Immunomodulatory Proteins using Combinatorial Plasmid Expression Cassettes

Multi-Modular Plasmid Expression Cassettes Demonstrate the Ability to Produce Multiple Immunomodulatory Proteins in Human Cells

Combinations of nucleic acids, encoding GOF type I IFN inducing variants and cytokines, were cloned into the pATI-1.75 vector, using two separate ORFs, under the control of the CMV and EF-1 a promoters (dual promoter system), or using T2A sequences within the ORFs and one promoter (single promoter system). The constructs also optionally included post-transcriptional regulatory elements (PREs), such as 3’ WPRE or HPRE, and/or polyadenylation signal sequences, such as the SV40 or bovine growth hormone (bGH) polyadenylation signals. The constructs prepared included those encoding the cytokines muIL-2, mu-IL-2 with codon optimization (muIL-2 CO), muIL-21, muIL-12p70, muIL-15Ra-IL-15sc, muIL-18, and muIFN-a2, and combinations thereof; the muSTING variant with the mutation R283G; and/or the murine co-stimulatory molecule 4-1BBL with a deletion of the cytoplasmic domain (mu4-lBBLAcyt). The sequences were confirmed by PCR.

To determine whether the combination plasmids, containing GOF type I IFN inducing variants, can induce functional type I IFN in human cells, the plasmids were assessed using HEK293T STING Null Reporter cells (InvivoGen), which do not contain endogenous STING. These cells express secreted embryonic alkaline phosphatase (SEAP), placed under the control of the endogenous IFN-stimulated response element (ISRE) promoter, where the coding sequence of ISRE has been replaced by the SEAP ORF using knock-in technology. Type I interferon activity can be assessed by monitoring type I IFN-stimulated SEAP production in the cell culture supernatants. In addition, supernatants were collected and evaluated for relative cytokine concentrations by ELISA.

To test the relative production of type I IFN and co-expressed cytokines, 2xl0 ⁵ 293T-Dual™ Null cells (InvivoGen) were plated one day prior on 24-well plates coated with poly-L-lysine, to achieve 80% confluency. On the day of transfection, 500 ng of plasmids encoding a panel of GOF variants, cytokines, and costimulatory molecules, alone, or in various combinations, were diluted in serum-free media and added to FuGENE® transfection reagent (Promega), at the proper reagent:DNA ratios. Cell culture supernatants from each sample were collected after overnight incubation, and 20 pL of the cell culture supernatants was added to 180 pL QUANTI-Blue™ reagent (InvivoGen). Type I interferon activation was determined by measuring ISRE-induced SEAP activity on a SpectraMax® M3 Spectrophotometer (Molecular Devices) at an absorbance wavelength of 650 nm. The muIL-2 constructs were evaluated for muIL-2 expression in a murine IL-2 ELISA (R&D Systems), according to the manufacturer’s instructions. The muIL-12p70 constructs were evaluated in a murine IL-12 ELISA (R&D Systems), according to the manufacturer’s recommendation. For the muIL-15Ra-IL-15sc constructs, the murine IL- 15 ELISA (eBioscience) was used, per kit instructions. IFN-a2 was measured using the RAW- Lucia™ ISG reporter cell line (InvivoGen). Murine IL-18 and murine IL-21 were measured by ELISA (Invitrogen).

As shown in the table below, assays performed to detect the presence of secreted functional proteins demonstrated high expression of multiple combination payloads. These include cytokine combinations, as well as combinations with type I IFN inducing GOF variants, and/or with co- stimulatory payloads. These data validate the ability of the platform provided herein to express multiple immunomodulatory proteins from a single expression cassette.

Constitutively Active Type I IFN Inducing Variants have Unique Cytokine and Chemokine Profiles

Supernatants harvested from HEK293T transfection experiments with a panel of human type I IFN inducing GOF variants, tested as described above, were evaluated for downstream signaling differences using a human anti-viral CBA panel (BD Biosciences), according to the manufacturer’s protocol. Each transfection was performed in duplicate, and cytokine levels were measured. The average of two measurements was calculated. Fold increase in average cytokine secretion was calculated compared to untransfected wells, in which the average cytokine secretion was set as 1.00.

As shown in the table below, low levels of human IL-12p70 were produced in the cells. However, several human type I IFN inducing GOF variants induced the production of type I IFN-a2 and/or IFN- , including the phosphomimetic IRF3 variant (huIRF3 S396D), as well as several constitutively active STING variants. A number of these GOF variants produced high levels of secreted CXCL10, demonstrating the ability of these expressed variants to recruit T-cells into the tumor microenvironment. The highest levels of CXCL10 expression were observed following the expression of the huSTING variant with the mutation R284G.

These data validate the use of multiplexed plasmid expression constructs, containing type I IFN inducing GOF variants, for induction of functional downstream cytokines and chemokines, such as CXCLlO/IP-10. These variants all have unique signatures, and STING GOF variants induced the highest levels of CXCL10 secretion, particularly huSTING R284G (corresponding to muSTING R283G).

Immune Cell Co-Culture Assays Identify Optimal Combinations of Immunomodulatory Targets

In order to determine the optimal combinations of cytokines to elicit T-cell recruitment and activation, a panel was tested in a macrophage and T-cell co-culture assay. Golden ticket (STING deficient) murine primary bone marrow-derived macrophages (BMMs) were generated as described above (see, Example 6), using a 24-well plate. Each well was transfected with the appropriate DNA constructs using FuGENE® transfection reagents (Promega). The constructs included those encoding muIL-2 CO, muIL-12p70, muSTING-R283G, muIL-2 CO + muIL-12p70, muIL- 15Ra-IL-15sc + muIL-12p70, and muIL-12p70 + muSTING-R283G + muIL-18. 24 hours post-transfection, 100 uL of cell culture supernatants were harvested from the wells for a flow cytometry-based cytokine bead array (CBA). In parallel, two spleens from C57BL/6 mice were dissected, and splenic CD4 ⁺ and CD8 ⁺ T-cells were isolated following instructions from a mouse T-cell isolation kit (StemCell Technologies). 200,000 isolated T-cells per well were then added to the transfected cells, with or without CD3s antibody (clone 145-2C11, BioLegend), at a final concentration of 0.5 pg/ml per well. At 24 and 48 hours post-addition of T-cells to the transfected cells, 100 pL of the co-culture supernatants were harvested from the wells for a flow-cytometry based cytokine bead array. Supernatants from transfected bone marrow macrophages (BMMs) and bone marrow macrophage/T-cell co-cultures were analyzed for their cytokine content, using murine anti-viral and murine Thl specific cytokine bead arrays, respectively.

As shown in the table below, only the plasmids encoding muSTING-R283G elicited CXCL10 production from the macrophages, while plasmids encoding muIL- 12p70, alone or in combination with other proteins, elicited the highest levels of IFNy from the co-cultured T-cells, which were greater than the background amount resulting from CD3e stimulation. The combination of muIL-12p70 + muIL-18 + muSTING-R283G was able to induce both CXCL10 production from macrophages, and IFNY production from co-cultured T-cells.

These data demonstrate the feasibility of expressing multiple immunomodulatory payloads from a single plasmid, as well as the synergistic activities of these combinations.

Combinatorial Immunotherapy Demonstrates Enhanced Anti-Tumor Activity in a Murine Model of Colorectal Adenocarcinoma In order to determine whether tumor-specific delivery of a combination of cytokines enhances anti-tumor efficacy, a construct encoding the combination of muIL-12p70, muIL-18, and muSTING-R283G (CMV-muIL-12p70_T2A_muIL-18 + EF-la-muSTING-R283G-WPRE), was assessed in the MC38 murine model of colorectal adenocarcinoma. For this study, 6-8 week-old female C57BL/6 mice (5 mice per group) were inoculated SC in the right flank with MC38 colorectal adenocarcinoma cells (5xlO ⁵ cells in 100 pL PBS). Mice bearing established flank tumors were IV injected on day 10 with IxlO ⁷ CFUs of strain YS1646Aas6//AFLG/A/?a' 7’/Aczn _J 8, containing a plasmid encoding the combination of muIL-12p70, muIL-18, and muSTING-R283G (CMV-muIL-12p70_T2A_muIL-18 + EF-la-muSTING-R283G-WPRE), or with PBS vehicle control.

The combination therapy was very well tolerated, with only an initial weight loss of 3.6% that had fully recovered 3 days later. This is in marked contrast to the toxicities observed with systemic administration of these cytokines (IL-12p70 and IL- 18). Compared to PBS control, the combination therapy was highly effective and curative (92.3% TGI, 60% cure rate, day 30). These data demonstrate the potency and safety of delivering combinations of cytokines in a tumor-specific manner, using the immunostimulatory bacterial strains described herein.

EXAMPLE 15

Protein Engineering Screening to Identify Improved Gain-of-Function Mutations in STING, RIG-I, MDA5, IRF3, IRF7, and other Interferon Pathway Genes

Gain-of-function (GOF) amino acid mutants that are constitutively active and that promote interferonopathies are identified from humans. Many GOF mutations occur due to single base pair nucleotide changes that alter the amino acid codon at that particular position in the gene. For example, in STING, the V147L mutation occurs due to a mutation at c.439G- C; N154S occurs due to a mutation at c.461 A->G; and V155M occurs due to a mutation at c.463G- A. The purpose of the screening was to identify constitutively active mutants that lead to high levels of type I interferon expression. Designed mutations, at sites known to promote interferonopathies when mutated in human patients, allow for a greater number of amino acid substitutions to be tested. In this example, site-directed mutagenesis with designed amino acids is performed at the positions of known mutations (see, Table below, listing mutations in genes that promote interferonopathies), to identify mutations with enhanced activity, that lead to high level type I interferon expression.

Amino acid residues R197, D205, R310, R293, T294, E296, S272, Q273, E316, D231, R232, K236, S358, E360, S366, and R238, with reference to the sequence of human STING, as set forth in SEQ ID N0s:305-309, correspond to amino acid residues R196, D204, R309, R292, T293, E295, S271, Q272, E315, D230, R231, K235, S357, E359, S365, and R237, respectively, with reference to the sequence of murine STING, as set forth in SEQ ID NO:369.

PCR primers are generated with designed substitutions flanked on the 5’ and 3’ ends with homologous cDNA sequences from the gene. The QuikChange® Site- Directed Mutagenesis kit (Agilent), or another comparable commercially available kit, is used to generate a PCR product incorporating the designed mutation. PCR amplified plasmids are treated with Dpnl, then electroporated into competent E. coll cells. Individual clones are isolated, plasmid mini-preps are performed, and the sequence identity of the desired mutation is confirmed. Larger scale plasmid preparations are then performed (using a Qiagen Kit), and the DNA is transfected into HEK293T STING Reporter cells (InvivoGen), which do not contain endogenous STING. These cells express Lucia™ luciferase, a secreted luciferase, placed under the control of the endogenous IFN-P promoter; the coding sequence of IFN-p has been replaced by the Lucia™ luciferase ORF using knock-in technology. Constitutively activate mutants then are identified and ranked by measurement of IFN-P promoter induced expression of luciferase activity.

EXAMPLE 16

Various Species of Immunostimulatory Bacteria with Inactivating Deletions Corresponding to the Deletions in Salmonella typhimurium

The Examples above describe exemplary modifications to the genome of Salmonella typhimurium to increase targeting to and accumulation of Salmonella typhimurium in immune-resident myeloid cells and in the tumor microenvironment, to deliver therapeutic products/payloads to tumors, and to reduce toxicity of the bacteria by eliminating their ability to infect other cell types. These genetic modifications similarly can be introduced into other bacterial strains and species, such as by deletion of the corresponding genes in other species, as described below.

Escherichia coli

In-frame chromosomal deletions of the IpxM, pnrM, asd,fliC,fliE,pagP, ansB and csgD genes are made sequentially in E. coli strains using a technique based on the recombineering methods described by Datsenko and Wanner (Proc. Natl. Acad. Sci. U.S.A. 97:6640-6645 (2000)). The genes in /■’. coli are as follows:

1) IpxM, encoding myristoyl-acyl carrier protein-dependent acyltransferase \E. coli strain K-12, substrain MG1655; NCBI Gene ID: 945143];

2) purM, encoding phosphoribosylformylglycinamide cyclo-ligase [E. coli strain K-12, substrain MG1655; NCBI Gene ID: 946975];

3) asd, encoding aspartate-semialdehyde dehydrogenase [//. coli strain K-12, substrain MG1655; NCBI Gene ID: 947939];

4) fliC, encoding flagellar filament structural protein \E. coli strain K-12, substrain MG1655; NCBI Gene ID: 949101];

5)fliE, encoding flagellar basal -body protein FliE \E. coli strain K-12, substrain MG1655; NCBI Gene ID: 946446];

6) pagP, encoding Lipid IVA palmitoyltransferase \E. coli strain K-12, substrain MG1655; NCBI Gene ID: 946360];

7) ansB, encoding L-asparaginase 2 [E. coli strain K-12, substrain MG1655; NCBI Gene ID: 947454];

8) csgD, encoding DNA-binding transcriptional dual regulator CsgD \E. coli strain K-12, substrain MG1655; NCBI Gene ID: 949119]; and

9) rpsM, encoding 30S ribosomal subunit protein S13 (promoter) coli strain K-12, substrain MG1655; NCBI Gene ID: 947791],

Briefly, a specific chromosomal sequence is replaced with a selectable antibiotic resistance marker flanked by homology arms, and is subsequently removed by a cre/loxP system. The 5’ and 3’ flanking sequences of each target gene are identified and cloned into a plasmid vector on opposing sides of an antibiotic resistance gene. The gene deletion cassette, containing the antibiotic resistance gene and flanking 5’ and 3’ homology arms is PCR amplified, gel purified, and introduced into the E. coli strain by electroporation. Electroporated cells are recovered, and transformants are selected for on antibiotic plates. The antibiotic marker is then cured using a cre/loxP recombination system, in which antibiotic resistant clones are transformed with a cre-expressing temperature-dependent plasmid. Colonies are selected at 30 °C and subsequently eliminated by serial passage at 42 °C, and then screened for loss of antibiotic resistance. Antibiotic sensitive clones are confirmed for gene deletion by colony PCR and sequence analysis.

Increasing Resistance or Rendering E. coli Resistant to Human Complement

Expression of the Salmonella typhimurium rck (resistance to complement killing) gene, the Yersinia enterocolitica homolog ail (attachment invasion locus), or the Salmonella typhimurium pgtE (outer membrane serine protease) gene, in the E. coli SlpxM! purMI SasdJ SfliCI SfliEI EpagPI AansB! Scsgl) strain is achieved by encoding the rck, ail, or pgtE gene sequence downstream from a constitutive promoter, such as E. coli or S. typhimurium rpsM, on a plasmid (asd complementation system compatible), or by insertion on the bacterial chromosome (at any of the IpxM, purM, asd,fliC,fliE,pagP, ansB, or csgD loci).

Salmonella typhi

In-frame chromosomal deletions of the msbB, purM, asd,fliC,flgB,pagP, ansB, and csgD genes are made sequentially in 5. typhi strains using the technique described above for A. co/z-based strains.

Expression of the Salmonella typhimurium rck (resistance to complement killing) gene, the Yersinia enterocolitica homolog ail, or the Salmonella typhimurium pgtE gene, in the 5. typhi EmsbBI EpurM/ Easd! EfliC! EflgB/ EpagPI EansB! EcsgD strain is achieved by encoding the rck, ail, or pgtE gene sequence downstream of a constitutive promoter, such as A typhimurium or S. typhi rpsM, on a plasmid (asd complementation system compatible), or by insertion on the bacterial chromosome (at any of the msbB, purM, asd,fliC,flgB, pagP, ansB, or csgD loci). The genes in S. typhi are as follows:

1) msbB (STY2097), encoding lipid A acyltransferase (Salmonella enterica subspecies enterica serovar Typhi, strain CT18; NCBI Gene ID: 1248440];

2) purM (STY2740), encoding phosphoribosylformylglycinamidine cycloligase (Salmonella enterica subspecies enterica serovar Typhi, strain CT 18, NCBI Gene ID: 1249054]; 3) asd (STY4271), encoding aspartate-semialdehyde dehydrogenase [Salmonella enterica subspecies enterica serovar Typhi, strain CT18; NCBI Gene ID: 1250488];

4) fliC (STY2167), encoding flagellin [Salmonella enterica subspecies enterica serovar Typhi, strain CT18; NCBI Gene ID: 1248507];

5) flgB (STY1213), encoding flagellar basal -body rod protein FlgB [Salmonella enterica subspecies enterica serovar Typhi, strain CT18; NCBI Gene ID: 1247617];

6) pagP (STY0677), encoding antimicrobial peptide resistance and lipid A acylation protein [Salmonella enterica subspecies enterica serovar Typhi, strain CT18; NCBI Gene ID: 1247137];

7) ansB (STY3259), encoding L-asparaginase [Salmonella enterica subspecies enterica serovar Typhi, strain CT18; NCBI Gene ID: 1249541];

8) csgD (STY1179), encoding regulatory protein CsgD [Salmonella enterica subspecies enterica serovar Typhi, strain CT18; NCBI Gene ID: 1247585]; and

9) rpsM (STY4380), encoding 30S ribosomal subunit protein S13 (promoter) [Salmonella enterica subspecies enterica serovar Typhi, strain CT18; NCBI Gene ID: 1250594],

Listeria monocytogenes

In-frame chromosomal deletions of the purA,purQ,purS, asd, flaA,fliC, flgB, and ansB genes in Listeria monocytogenes is achieved by an allelic exchange technique using a temperature-sensitive shuttle vector, such as pKSV7, that confers antibiotic resistance and allows replication at low temperatures (30 °C), but is unable to replicate at higher temperature (43 °C). The 5’ and 3’ flanking sequences of each target gene are identified and cloned in tandem into the pKSV7 vector, which is transformed into recipient Listeria monocytogenes and selected for by plating on agar plates with antibiotic. Chromosomal integration of the plasmid is induced by serial passage of antibiotic-resistant transformants at 42 °C under selection. Subsequent sequential subculturing of the strain at 30 °C results in subpopulations of cells in which the plasmids are excised through a second crossover event, producing reversions to the original wild-type gene, or incorporation of the 5’ and 3’ flanking sequence homology arms to generate the targeted deletion mutant. Antibiotic sensitive clones are screened at this step by colony PCR and sequence analysis. To increase complement resistance, expression of the Salmonella typhimurium rck (resistance to complement killing) gene, the Yersinia enterocolitica homolog ail, or the Salmonella typhimurium pgtE gene in the Listeria monocytogenes

Spur Al SpurOi SpurS/ asdl SflaA! SfliC! Sflg I ansB strain is achieved by encoding the rck, ail, or pgtE gene sequence downstream of a constitutive promoter, such as Phyp er or Pheiper, on a plasmid (asd complementation system compatible), or by insertion on the bacterial chromosome (at any of the purA,purQ,purS, asd,flaA,fliC, flgB, or ansB loci). The genes in L. monocytogenes are as follows:

1) purA (lmo0055), encoding adenylosuccinate synthetase [Listeria monocytogenes strain EGD-e; NCBI Gene ID: 986069];

2) purQ (lmol769), encoding phosphoribosylformylglycinamidine synthase II [Listeria monocytogenes strain EGD-e; NCBI Gene ID: 985972];

3) purS (Imo 1771), encoding phosphoribosylformylglycinamidine synthase subunit PurS [Listeria monocytogenes strain EGD-e; NCBI Gene ID: 985970];

4) asd (Imo 1437), encoding aspartate-semialdehyde dehydrogenase [Listeria monocytogenes strain EGD-e; NCBI Gene ID: 986492];

5) flaA (lmo0690), encoding flagellin [Listeria monocytogenes strain EGD-e; NCBI Gene ID: 987167];

6) fliE (lmo0712), encoding flagellar hook-basal body protein FliE [ Listeria monocytogenes strain EGD-e; NCBI Gene ID: 985062];

7) flgB (hno0710), encoding flagellar basal -body rod protein FlgB [Listeria monocytogenes strain EGD-e; NCBI Gene ID 985059]; and

8) ansB (Imo 1663), encoding asparagine synthetase [Listeria monocytogenes strain EGD-e; NCBI Gene ID: 985663],

The genes msbB and pagP are absent in Listeria monocytogenes, which is a Gram-positive bacterium. CsgD also is absent in Listeria monocytogenes. Instead, Listeria express lep, encoding the Listeria cellulose binding protein that is involved in biofilm formation, which also can be deleted.

Bifidobacterium longum

In-frame chromosomal deletions of BL1122 (purM), BL0492 (asd) and BL1142 (encoding an L-asparaginase precursor) in Bifidobacterium longum is achieved by an allelic exchange technique using an incompatible plasmid vector system, in which a conditional -replication vector, such as pBS423-A/'c7?4, which lacks the plasmid replication gene, repA, is initially integrated into the genome, providing antibiotic resistance. A second plasmid, such as pTBRIOl-CM, encoding the rep A gene, is subsequently transformed and facilitates a second crossover event that selects for excision of the initial integrant. The 5’ and 3’ flanking sequences of each target gene are identified and cloned in tandem into the conditional replication vector, lacking repA, and transformed into the Bifidobacterium longum

ABL1122/ABL0492/ABL1142 strain. Crossover recombination events can occur at homology arm sequences, and successful plasmid integrants are selected for and isolated on antibiotic plates. Integrants are then transformed with an incompatible plasmid encoding a functional copy of repA, which facilitates a second crossover event and excision of the repA deficient plasmid, which is subsequently lost due to plasmid incompatibility. Genomic deletions are confirmed by colony PCR and sequence analysis, and the remaining plasmid is cured by removing selection and subsequent Rif treatment.

Expression of the Salmonella typhimurium rck (resistance to complement killing) gene, the Yersinia enterocolitica homolog ail, or the Salmonella typhimurium pgtE gene in the Bifidobacterium longum ABL1122/ABL0492/ABL1142 strain is achieved by encoding the rck, ail, or pgtE gene sequence downstream of a strong constitutive promoter, such as P _gap , on a plasmid (asd complementation system compatible), or by insertion on the bacterial chromosome (at any of the BL 1122, BL0492, or BL1142 loci). The Bifidobacterium longum genes are as follows:

1) purM (BL1122), encoding phosphoribosylformylglycinamidine cyclo-ligase [Bifidobacterium longum strain NCC2705; NCBI Gene ID: 1022669];

2) asd (BL0492), encoding aspartate-semialdehyde dehydrogenase [Bifidobacterium longum strain NCC2705; NCBI Gene ID: 1023089];

3) BL1142, encoding an L-asparaginase precursor (Ntn_Asparaginase_2_like; L-Asparaginase type 2-like enzymes of the NTN-hydrolase superfamily) [Bifidobacterium longum strain NCC2705; NCBI Gene ID: 1023120]; and

4) BL1363 gap (promoter) [Bifidobacterium longum strain NCC2705; NCBI Gene ID: 1022828],

Bifidobacterium longum are non-motile and lack flagellin, and are Grampositive and lack msbB and pagP. The ansB gene is present, but encodes aspartate ammonia-lyase, which catalyzes the formation of fumarate from aspartate (aspA/ansB) [BL0338, Bifidobacterium longum strain NCC2705; NCBI Gene ID: 1023259],

Clostridium novyi

In-frame chromosomal deletions of NT01CX_RS09765, NT01CX_RS07625, and NT01CX_RS04325 (asd) the flagellin genes NT01CX_RS04995, NT01CX_RS04990, NT01CX_RS05070, and NT01CX_RS05075; and the flagellar basal body rod protein genes NT01CX_RS05080 flgBfi NT01CX_RS05085 flgC), and NT01CX_RS05215 (flgG) in Clostridium is achieved by an allelic exchange technique requiring a counter-selection method including toxin-antitoxin systems, which requires an inducible promoter and toxic gene, such as the E. coli mRNA interferase mazF. The 5’ and 3’ flanking sequences of each target gene are identified and cloned in a configuration on opposing sides of a frt-flanked antibiotic resistance cassette in an allelic exchange counter-selection-containing vector, and transformed into Clostridium novyi. The mazF gene is encoded under the control of an inducible lac promoter on the allelic exchange vector, and permits selection of a double crossover event by growth on lactose-supplemented agar plates. Genomic deletions are confirmed by colony PCR and sequence analysis. Flp-frt recombination can then be used to cure the antibiotic resistance cassette from the chromosome.

Expression of the Salmonella typhimurium rck (resistance to complement killing) gene, the Yersinia enterocolitica homolog ail, or the Salmonella typhimurium pgtE gene in the Clostridium novyi ANT01CX_RS09765/A NT01CX_RS07625/ANT01CX_RS04325/ANT01CX_RS04995/ANT01CX_RS049 90 /ANTO 1 CX_RS05070/ANT01 CX_RS05075/ANT01 CX_RS05080/ANT01 CX_RS050 85/ANT01CX_RS05215 strain is achieved by encoding the rck, ail, or pgtE gene sequence downstream from a strong constitutive promoter, such as Pthi, P _p tb, or other variants, on a plasmid (asd complementation system compatible), or by insertion on the bacterial chromosome (at any of the NT01CX_RS09765, NT01CX_RS07625, NT01CX_RS04325, NT01CX_RS04995, NT01CX_RS04990, NT01CX_RS05070, NT01CX_RS05075, NT01CX_RS05080, NT01CX_RS05085, or NT01CX_RS05215 loci). The Clostridium novyi genes are as follows:

1) NT01CX_RS09765, encoding AIR synthase [Clostridium novyi strain NT; NCBI Gene ID: 4541583]; 2) NT01CX_RS07625, encoding phosphonbosylformylglycinamidine cycloligase [Clostridium novyi strain NT; NCBI Gene ID: 4540669];

3) NT01CX_RS04325 (asd), encoding aspartate-semialdehyde dehydrogenase [Clostridium novyi strain NT; NCBI Gene ID: 4541762];

4) NT01CX_RS04995, encoding flagellin [Clostridium novyi strain NT; NCBI Gene ID: 4541703];

5) NT01CX_RS04990, encoding flagellin [Clostridium novyi strain NT; NCBI Gene ID: 4539984];

6) NT01CX_RS05070, encoding flagellin [Clostridium novyi NT; NCBI Gene ID: 4539886];

7) NT01CX_RS05075, encoding flagellin [Clostridium novyi NT; NCBI Gene ID: 4539699];

8) NT01CX_RS05080 (flg ), encoding flagellar basal body rod protein FlgB [Clostridium novyi strain NT; NCBI Gene ID: 4540637];

9) NT01CX_RS05085 (flgC encoding flagellar basal body rod protein FlgC [Clostridium novyi strain NT; NCBI Gene ID: 4540143]; and

10) NT01CX_RS05215 (flgG), encoding flagellar basal body rod protein FlgG [Clostridium novyi strain NT; NCBI Gene ID: 4540245],

Clostridium novyi is Gram-positive, and lacks msbB and pagP.

EXAMPLE 17

Immunostimulatory Bacteria Modified to Express Vertebrate STING Variants that Induce Stronger Type I IFN Signaling and/or Weaker NF-KB Signaling than Human STING

STING signaling activates two signaling pathways. The first is the TANK binding kinase 1 (TBK1)/IRF3 axis, resulting in the induction of type I IFNs, and the activation of dendritic cells (DCs) and cross-presentation of tumor antigens to activate CD8 ⁺ T-cell mediated anti-tumor immunity. The second is the nuclear factor kappalight-chain-enhancer of activated B-cell (NF-KB) signaling axis, resulting in a pro- inflammatory response, but not in the activation of the DCs and CD8 ⁺ T-cells that are required for anti-tumor immunity. Bacterially-based cancer immunotherapies are limited in their ability to induce type I IFN to recruit and activate the CD8 ⁺ T-cells that are necessary to promote tumor antigen cross-presentation and durable anti-tumor immunity. Hence, provided are immunostimulatory bacteria herein that induce and/or increase type I IFN signaling, and that have decreased NF-KB signaling, thereby increasing the induction of CD8 T-cell mediated anti -tumor immunity, and enhancing the therapeutic efficacy of the bacteria. The immunostimulatory bacteria described above encode modified STING proteins that are gain-of-function mutants of STING that can increase induction of type I IFN compared to wild-type STING, or render the expression of type I IFN constitutive. In this example (and also described in the detailed description), the STING protein is modified to reduce or eliminate NF- KB signaling activity, and to retain the ability to induce type I IFN, and/or is modified for increased or constitutive type I IFN expression. This results in immunostimulatory bacteria that induce anti -tumor immunity, and do not induce (or induce less) NF-KB signaling that normally results from infection by bacterial pathogens.

STING proteins from different species exhibit different levels of type I IFN and NF-KB signaling activities. For example, STING signaling in human and mouse cells results in a strong type I IFN response, and a weak pro-inflammatory NF-KB response. STING signaling in ray-finned fish, such as salmon and zebrafish, in comparison, elicits robust activation of a primarily NF-KB-driven response, that is more than 100-fold higher compared with the IRF3 -driven (i.e., type I IFN inducing) response. In other species, such as Tasmanian devil, STING signaling results in a type I IFN response, but essentially no NF-KB response. The immunostimulatory bacteria provided herein encode STING from non-human species, such as Tasmanian devil STING, in order to exploit the ability of STING to induce a type I IFN response, but without the concomitant induction of an NF-KB response. As described herein, these non-human STING proteins also are modified by mutation to increase the type I IFN response, or to render it constitutive. The identified mutations that have this effect in human STING are introduced into the non-human STING proteins. The corresponding residues are identified by alignment.

Also provided are chimeras in which the C-terminal tail (CTT) of STING is replaced in one species, such as human, with the CTT from a second (e.g., non- human) species STING protein that exhibits little or no NF-KB signaling activity. The CTT is an unstructured stretch of approximately 40 amino acids that contains sequence motifs required for STING phosphorylation and recruitment of IRF3. It can shape downstream immunity by altering the balance between type I IFN and NF-KB signaling. This is controlled through independent modules in the CTT, including IRF3-, TBK1-, and TRAF6-binding modules. For example, human STING residue S366 (see, e.g., SEQ ID N0s:305-309) is a primary TBK1 phosphorylation site that is part of an LxIS motif in the CTT, which is required for IRF3 binding, while a second PxPLR motif, including residue L374, is required for TBK1 binding. The LxIS and PxPLR motifs are highly conserved in all vertebrate STING alleles. Replacing the CTT of human STING with that of, for example, Tasmanian devil STING, produces a STING variant that induces a type I IFN response, but not an NF-KB response.

In this Example, the immunostimulatory bacteria are engineered to express a STING variant with increased type I IFN signaling, and/or reduced NF-KB signaling, compared to wild-type (WT) human STING (see, e.g., SEQ ID N0s:305-309). The STING variants can be from a non-human vertebrate, such as a mammalian, bird, reptilian, amphibian, or fish species. Species from which the non-human STING proteins are derived include, but are not limited to, Tasmanian devil (Sarcophilus harrisii, SEQ ID NO:349), marmoset (Callithrix jacchus,' SEQ ID NO:359), cattle (Bos taurus; SEQ ID NO:360), cat (Felis catus,' SEQ ID NO:356), ostrich (Struthio camelus australis,' SEQ ID NO:361), crested ibis (Nipponia nipporr, SEQ ID NO:362), coelacanth (Latimeria chalumnae; SEQ ID NOs:363-364), boar (Sus scrofcr, SEQ ID NO: 365), bat (Rousettus aegyptiacus,' SEQ ID NO: 366), manatee (Trichechus manatus latirostris, SEQ ID NO: 367), ghost shark (Callorhinchus milii, SEQ ID NO:368), and mouse (Mus musculus,' SEQ ID NO:369). These vertebrate STING proteins readily activate immune signaling in human cells, indicating that the molecular mechanism of STING signaling is shared in vertebrates (see, e.g., de Oliveira Mann et al. (2019) Cell Reports 27:1165-1175). STING proteins from these species induce less NF-KB signal activation and/or more type I IFN signal activation, than human STING (see, e.g., de Oliveira Mann et al. (2019), Fig. 1A). Wild-type or modified STING proteins from different non-human species can be expressed by the immunostimulatory bacteria herein, as can chimeras of human and non-human STING proteins.

The various non-human STING proteins are modified, such that the non- human STING has lower NF-KB activation, and, optionally, higher type I interferon activation, than human STING. These non-human STING proteins are modified to include a mutation or mutations so that they have increased type I IFN activity, or act constitutively, in the absence of cytosolic nucleic acid ligands (e.g., CDNs). The mutations typically are amino acid mutations, such as gain-of-function mutations, that are associated with interferonopathies in humans. The corresponding mutations are introduced into the non-human species STING proteins, where corresponding amino acid residues are identified by alignment. For example, mutations include, but are not limited to, S102P, V147L, V147M, N154S, V155M, G166E, C206Y, G207E, S102P/F279L, F279L, R281Q, R284G, R284S, R284M, R284K, R284T, R197A, D205A, R310A, R293A, T294A, E296A, R197A/D205A, S272A/Q273A, R310A/E316A, E316A, E316N, E316Q, S272A, R293A/T294A/E296A, D231A, R232A, K236A, Q273A, S358A/E360A/S366A, D231A/R232A/K236A/R238A, S358A, E360A, S366A, R238A, R375A, and S324A/S326A, with reference to the sequence of human STING, as set forth in SEQ ID N0s:305-309. Corresponding mutations in STING from other species are listed in the tables below. The resulting variants of the non-human STING proteins include one or more of these mutations, and optionally, a CTT replacement, and optionally, a deletion in the TRAF6 binding site.

The STING variants include one or more replacements of the amino acid serine (S) or threonine (T) at a phosphorylation site, with aspartic acid (D), which is phosphomimetic, resulting in increased or constitutive activity. Other mutations include deletion or replacement of a phosphorylation site or sites, such as 324-326 SLS — ► ALA in STING, and other replacements to eliminate a phosphorylation site to reduce NF-KB signaling in STING. Additionally, chimeras of human STING with STING from other species are provided, in which the C-terminal tail (CTT) of human STING is replaced with the CTT of STING from another species that has lower NF- KB signaling activity, and/or higher type I IFN signaling activity. The variant STING proteins can include a deletion in the TRAF6 binding site of the CTT, to reduce NF-

KB signaling.

For example, modified STING variants include Tasmanian devil STING with the mutations C206Y (SEQ ID NO:350), or R284G (SEQ ID NO:351); a variant in which the CTT of human STING is replaced with the CTT of Tasmanian devil STING (SEQ ID NO:352); human STING with the mutation C206Y (SEQ ID NO:353), or R284G (SEQ ID NO:354), and where the CTT is replaced with the CTT of Tasmanian devil STING; wild-type human STING with a deletion in the TRAF6 binding domain (corresponding to residues 377-379 (DFS)) (SEQ ID NO:355); cat STING with the mutations C205Y (SEQ ID NO:357), or R283G (SEQ ID NO:358); and other such modified STING variants.

To determine the corresponding amino acid residues for the STING mutations, the wild-type STING protein sequences from various non-human species each were aligned with the wild-type human STING protein sequence (of the allelic variants of SEQ ID NO:305 (R232 allele) or SEQ ID NO:306 (H232 allele)). The alignments were performed using the Kalign sequence alignment tool, available from ebi.ac.uk/Tools/msa/kalign/, or the EMBOSS needle sequence alignment tool, available from ebi.ac.uk/Tools/psa/emboss_needle/. Figures 2-14 depict exemplary sequence alignments for human STING compared to STING proteins from Tasmanian devil, marmoset, cattle, cat, ostrich, crested ibis, coelacanth, zebrafish, boar, bat, manatee, ghost shark, and mouse species, respectively.

STING GOF Hybrid Variants Demonstrate Significantly Enhanced Type I Interferon to NF-KB Activity Ratios in Dendritic Cells

In order to determine the optimal STING GOF mutant that would elicit the highest levels of the CD8 ⁺ T-cell chemokine CXCL10 in mice, a panel of STING mutants were tested in murine primary bone marrow-derived dendritic cells (BMDCs). These included the Tasmanian devil STING with the constitutively active human GOF mutations C206Y (tazSTING C206Y; SEQ ID NO:350), or R284G (tazSTING R284G; SEQ ID NO:351); murine STING with the constitutive GOF mutants C205Y (muSTING C205Y), or R283G (muSTING R283G); cat STING with the constitutive GOF mutants C205Y (catSTING C205Y; SEQ ID NO:357), or R283G (catSTING R283G; SEQ ID NO:358); as well as variants in which the CTT of human STING (SEQ ID NO:370) was replaced with the CTT of Tasmanian devil STING (SEQ ID NO:371), and containing either wild-type human STING (huSTING tazCTT; see, e.g., SEQ ID NO:352), or the human STING GOF mutations C206Y (huSTING C206Y tazCTT; see, e.g., SEQ ID NO:353), or R284G (huSTING R284G tazCTT; see, e.g., SEQ ID NO:354). Also included were human STING variants with the constitutive, GOF mutations C206Y (huSTING C206Y), or R284G (huSTING R284G).

To test these, murine bone marrow was isolated and flushed into 1.5 mL Eppendorf tubes, and spun at 1200 RPM for 5 minutes, to collect the bone marrow cells. Cells were washed once in RPMI-1640 + 10% FBS, then seeded in 96-well TC- treated plates in RPMI-1640 + 10% FBS with 20 ng/ml GM-CSF. Every 2 days, 50% of the medium was replaced with fresh complete media. After six days, non-adherent cells were pipetted off the wells and re-seeded at le5 cells per well in RPMI-1640 + 10% FBS in a 96-well plate for transfection. Cells were transfected using Viromer® RED, according to the manufacturer’s instructions. Briefly, 200 ng of plasmid DNA from a panel of STING GOF mutants, as well as untransfected control, were diluted in the provided buffer, and mixed with 0.08 pL of Viromer® RED and incubated at room temperature for 15 minutes to allow the DNA/Viromer® RED complexes to form. The DNA/Viromer® RED complexes were then slowly added to each well of the 96-well plate (in duplicates), and the plate was incubated at 37 °C in a CO2 incubator. Supernatants were harvested at 48 hours, and assayed for murine CXCL10 (IP- 10) using a flow cytometry -based cytokine bead array (CBA), according to the manufacturer’s protocol.

As shown in the table below, the construct that induced the highest expression of murine CXCL10 was human STING containing the GOF mutation R284G, and containing a replacement of the CTT of human STING with the CTT of Tasmanian devil STING (huSTING R284G tazCTT). The next highest expression of CXCL10 was induced by the human STING variant containing the GOF mutation C206Y (huSTING C206Y), and the Tasmanian devil STING construct containing the human STING GOF mutation R284G (tazSTING R284G). The human STING GOF mutants (huSTING C206Y and huSTING R284G) were more potent than the corresponding murine STING GOF mutants (muSTING C205Y and muSTING R283G, respectively), which were even less potent than the cat STING mutants containing the same GOF mutations (catSTING C205Y and catSTINGR283G, respectively), in primary murine dendritic cells. These data demonstrate that STING proteins obtained from other species, such as Tasmanian devil, can be combined with constitutive GOF human STING mutations, to elicit potent T-cell recruiting chemokines.

STING GOF Hybrid Variants Demonstrate Significantly Enhanced Type I Interferon to NF-KB Activity Ratios in Human Monocytes

In order to demonstrate that the ratio of STING-induced type I interferon to NF-KB signaling can be altered using STING GOF hybrid variants from other species, a panel was tested in a human monocyte cell line. The panel included wild-type human STING (huSTING), and huSTING mutants with the constitutive human GOF mutations C206Y, or R284G; wild-type Tasmanian devil STING (tazSTING), and tazSTING mutants with the constitutive GOF mutations C206Y, or R284G; wild-type cat STING, and catSTING mutants with the constitutive GOF mutations C205, or R283G; murine STING mutants with the constitutive GOF mutations C205Y, or R283G; and the variants in which the CTT of human STING was replaced with the CTT of Tasmanian devil STING, and containing either wild-type human STING, or the human GOF STING mutations C206Y, or R284G. Also included were a wild-type human STING with a deletion in the TRAF6 binding domain (corresponding to residues 377-379, DFS, see, e.g., SEQ ID NO:355), and wild-type zebrafish STING.

For this experiment, the THPl-Dual™ KO STING cells were utilized, which have been altered to lack endogenous STING, and to also express Lucia™ luciferase, a secreted luciferase, placed under the control of the endogenous IFN-P promoter. Constitutively active STING GOF mutants then were identified, and ranked by measurement of IFN-P promoter induced expression of luciferase activity. These cells also express secreted embryonic alkaline phosphatase (SEAP), placed under the control of the endogenous NF-KB promoter, where the coding sequence of NF-KB has been replaced by the SEAP ORF using knock-in technology. NF-KB activity induced by STING GOF mutants can be assessed by monitoring SEAP production in the cell supernatants.

For this experiment, THPl-Dual™ KO STING cells were transfected using Viromer® RED, according to the manufacturer’s instructions. Briefly, 200 ng of plasmid DNA from a panel of STING GOF mutants, as well as untransfected control, were diluted in the provided buffer, and mixed with 0.08 pL of Viromer® RED and incubated at room temperature for 15 minutes to allow the DNA/Viromer® RED complexes to form. The DNA/Viromer® RED complexes were then slowly added to each well of the 96-well plate (in duplicates), and the plate was incubated at 37 °C in a CO2 incubator. In addition, the wild-type STING variants were treated with or without the STING agonist 3’5’ RpRp c-di-AMP (CDN, InvivoGen), an analog of the clinical compound ADU-S100, which was added to the cells after 24 hours of incubation at 10 pg/mL. Supernatants were harvested at 48 hours, and assayed for NF-KB-SEAP and IFN-Lucia reporter signals, according to the manufacturer’s protocol. Briefly, 10 pL of the cell culture supernatants was added to 50 pL QUANTI-Blue™ reagent (InvivoGen) (which is used for measuring SEAP). NF-KB activation was determined by measuring NF-KB -induced SEAP activity on a SpectraMax® M3 Spectrophotometer (Molecular Devices), at an absorbance (Abs) wavelength of 650 nm. For measuring type I interferon activity from IFN-Lucia, 10 pL of the cell culture supernatants was added to 50 pL QUANTI-Luc™, containing the coelenterazine substrate for the luciferase reaction, which produces a light signal that is quantified using a SpectraMax® M3 luminometer, and expressed as relative light units (RLUs).

As shown in the table below, the highest type I IFN responses were observed from the variant in which the CTT of human STING was replaced with the CTT of Tasmanian devil STING, and that contained the human STING GOF mutation R284G (huSTING R284G tazCTT), as well as from the wild-type zebrafish STING with the CDN STING agonist (zfSTING WT + CDN). However, unlike the wild-type zebrafish STING, which had very high NF-KB signaling, the huSTING R284G tazCTT variant had high type I IFN signaling with much lower NF-KB signaling activity. The best ratio of higher type I IFN to lower NF-KB signaling was found with the Tasmanian devil STING variant containing the human STING GOF mutation R284G (tazSTING R284G).

SD = Standard deviation

These data further demonstrate using non-human STING proteins, such as the STING protein from Tasmanian devil, and combining them with human constitutive gain-of-function STING mutations, in order to enhance the beneficial type I interferon activity, while minimizing the immunosuppressive NF-KB activity, in human monocytes.

EXAMPLE 18

A. typhimurium Lipoprotein Knockout by Deletion of the IppA and IppB Genes The live attenuated S. typhimurium YS1646 strain, containing the asdi AFI.G/ XpagP! AansB! csgD gene deletions, was engineered to delete the IppA (SEQ ID NO:387) and IppB (SEQ ID NO:388) genes, in order to remove membrane surface lipoproteins. This reduces pro-inflammatory TLR2 activation, which reduces immunosuppressive cytokines and improves anti-tumor adaptive immunity. As shown below, this also enhances plasmid delivery and encoded protein expression in the tumor.

Strain Engineering and Characterization

Deletion of IppA Gene

The IppA gene was deleted from the chromosome of the

YS1646Aa«// AFLG/ NpagP/ XansBI XcsgD strain using modifications of the method of Datsenko and Wanner (Proc. Natl. Acad. Sci. U.S.A. 97:6640-6645 (2000)), as described in detail above. Synthetic IppA gene homology arm sequences that contained 231 and 200 bases of the left hand and right hand sequence, respectively, flanking the IppA gene, were synthesized and cloned into a plasmid called pSL0148 (SEQ ID NO:231). The sequence for the IppA gene is shown in SEQ ID NO:387; appropriate primers for PCR amplification were designed using the gene sequence. A kanamycin gene cassette flanked by cre/loxP sites then was cloned into plasmid pSL0148, and the IppA gene knockout cassette was then PCR amplified with primers IppA-l and lppA-2, gel purified, and then introduced into the YS I646Ac/.st/AFLG/A/?c/g/ ^> /A<:7//.sA/Ac ^, .sg/J strain carrying the temperature sensitive lambda red recombination plasmid pKD46, by electroporation. Electroporated cells were recovered in SOC + DAP medium, and plated onto LB agar plates supplemented with kanamycin (20 pg/mL) and diaminopimelic acid (DAP, 50 pg/mL). Colonies were selected and screened for insertion of the knockout fragment by PCR using primers IppA- 3 and lppA-4. pKD46 then was cured by culturing the selected kanamycin resistant strain at 42 °C and screening for loss of ampicillin resistance. The kanamycin resistance marker then was cured by electroporation of a temperaturesensitive plasmid expressing the Cre recombinase (pJW168), and Amp ^R colonies were selected at 30 °C; pJW168 was subsequently eliminated by growing cultures at 42 °C. Selected IppA knockout clones were then tested for loss of the kanamycin marker by PCR, using primers flanking the sites of disruption (lppA-3 and lppA-4), and evaluation of the electrophoretic mobility was performed on agarose gels. This mutant derivative of strain YS1646 was designated

YS 1646 kasd! \¥LGI\pagPl^nsBlkcsgDI MppA .

Deletion of IppB Gene

The IppB gene then was deleted in the YS1646Aa«Z/AFLG/A/?agP/Aaft£S/AcsgD/A//Y>A strain using modifications of the methods described above. Synthetic IppB gene homology arm sequences that contained 224 and 231 bases of the left hand and right hand sequence, respectively, flanking the IppB gene, were synthesized and cloned into a plasmid called pSL0148 (SEQ ID NO:231). The sequence for the IppB gene is shown in SEQ ID NO:388; appropriate primers for PCR amplification were designed using the gene sequence. A kanamycin gene cassette flanked by cre/loxP sites then was cloned into plasmid pSL0148, and the IppB gene knockout cassette was PCR amplified with primers IppB- 5 and lppB-6, gel purified, and introduced into strain

YS I646A«.szZ ^/ AFLG/ kpagPI kansB/kcsgDI AJppA, carrying the temperature-sensitive lambda red recombination plasmid pKD46, by electroporation. The kanamycin resistance gene then was cured by Cre-mediated recombination as described above, and the temperature-sensitive plasmids were cured by growth at non-permissive temperature. The IppA and IppB gene knockout sequences were amplified by PCR, using primers designated lppA-3 and lppA-4, and lppB-7 and lppB-8, respectively, and verified by DNA sequencing. This mutant derivative of strain YS1646 was designated YS I646Ar/ / AFLG/ BpagP! BansBI BcsgD! BlppAB, or nicknamed YS1646 //?/?A5.

In vitro Characterization of Engineered S. typhimurium Lipoprotein Knockout Strain

The YS 1646Aa.st/ ^/ AFLG/ BpagP! Ban B! Bcsgl!)! BlppAB strain, harboring the deletions of IppA and IppB, was evaluated for growth by overnight cultures in LB. Growth was measured using a SpectraMax® M3 Spectrophotometer (Molecular Devices) at 37 °C, reading the ODgoo every 15 minutes. The results demonstrated that strain YS1646 Basel! AFLG/ BpagP! BansB! BcsgD! Blpp AB was able to replicate in LB at a growth rate comparable to the parental YS I646A .sc// AFLG/ BpagP! BansBI BcsgD strain. These data demonstrate that the elimination of lipoprotein does not decrease the fitness of S. typhimurium in vitro.

Lipoprotein Deletion Enhances Plasmid Delivery to the Tumor After Systemic Administration

As TLR2 is expressed on vascular endothelial cells, and activation of TLR2 enhances vascular permeability, the effect of the BlppAB modification, and the subsequent reduction in TLR2 agonism, on tumor colonization was assessed. The effect on payload expression also was assessed. As shown below, while colonization of tumors was somewhat reduced, plasmid delivery and encoded gene expression was significantly increased, following systemic administration.

To demonstrate the impact of the lipoprotein knockout strains in a murine model of triple-negative breast cancer, 6-8 week-old female BALB/c mice (4 mice per group) were inoculated orthotopically in the 4 ^th mammary fat pad with EMT6 cells (5xl0 ⁵ cells in 100 pL PBS). Mice bearing 10-day established flank tumors were IV injected with a single dose of 1 xlO ⁷ CFUs of the YS l646Aa.st/ ^/ AFLG/ BpagP! BansB! Bcsgl)! BlppAB strain, or the parental YS I646Aa.st/ ^/ AFLG/ BpagP! BansBI BcsgD strain, each containing a plasmid encoding a luciferase protein (NanoLuciferase® luciferase, or NanoLuc®; Promega) that is secreted (secNanoLuc®), under the control of the CMV promoter. At day 7 post IV dosing, mice were euthanized, and tumors were homogenized and plated on LB plates, to enumerate the number of colony forming units (CFUs) per gram of tumor tissue. The parental YS1646AaswAFLG/A/ragP/Aan£S/AcsgD strain colonized tumors at a mean of 3.3 xlO ⁶ CFUs per gram of tumor tissue, while the lipoprotein- deleted YS I 646Aas<7/ AFI.G/ SpagPISansBI \csgDI \lppAB strain colonized the tumors with a 2-fold decreased mean of 1.18 x 10 ⁶ CFUs/g of tumor tissue.

In order to measure plasmid delivery to the tumor, and subsequent heterologous gene expression and protein secretion, the activity of the secNanoLuc® was measured. For this, homogenized tumors were assessed for luciferase activity using the NanoGio® detection reagent (Promega), and read on a SpectraMax® M3 Spectrophotometer/Luminometer (Molecular Devices). While the parental Y S I 646 Sasdl AFLG / SpagP '/ SansBI ScsgD strain induced average luminescence relative light units (RLUs) of 4482.6, the YS1646Aas6?/AFLG/A/?ag7LAans/LAcsg£>/A//?/M.8 strain induced a nearly 10-fold increased average of 33,926.6 RLUs. These data demonstrate the ability of the lipoprotein-deleted strain to improve tumor colonization, and to enhance payload expression.

These data reveal that, while deletion of lipoprotein somewhat reduces tumor colonization after IV dosing, it significantly enhances plasmid delivery and payload expression in the tumor. These data demonstrate that, contrary to the expectation from the art that deletion of these genes will decrease colonization, lipoprotein deletion enhances plasmid delivery and protein expression in the tumor microenvironment, and in tumors.

EXAMPLE 19

Salmonella Full purl Clean Deletion Gene Knockout Strain Engineering and Characterization

The purl gene in strain YS1646 (VNP20009) was not deleted; it was disrupted by a transposon (TnlO) insertion that resulted in a 16.6 kbp (kilobase pair) genomic inversion event, whereby two insertion sequence (IS) elements were subsequently incorporated into the genome, one within the purl (purM) gene, and the other 16.6 kbp upstream, in the intergenic region directly flanking the 3 ’-end of the acrD gene. The region between the two IS elements is inverted and contains 18 genes, including yffB. DC51_2568, upp, uraA, yfgE, yfgD, DC51_2573, per Pl, purC, and others. The insertion sequence element in the intergenic region encodes a fully functional transposase (see, e.g, Broadway etal. (2014) J Biotechnology 192: 177-178). The presence of such a transposase in a therapeutic strain presents a possible genetic stability concern. These elements were removed from the strain, for development of the strain as a human therapy. The presence of the complete genetic sequence of the purl gene, disrupted by means of a chromosomal reengagement, leaves open the possibility of reversion to a wild-type gene.

The msbB gene in strain YS1646 also was not fully deleted, but was disrupted by a genetically engineered 511 bp deletion (of the 972 bp gene) that resulted in an extension of the pykA gene (encoding pyruvate kinase), replacing the last 5 amino acid codons with 13 new codons (see, e.g., Broadway et al. (2014) J. Biotechnology 192:177-178).

Strain YS1646Aasz// AFLG/ NpagP! ansB! csgD was modified to delete the remaining portions of the purl gene and the two transposon-associated insertion sequence elements present in strain YS1646. It is shown below, that with a full knockout of purl, the resulting bacterial cells have higher viability than the cells in which the gene is inactivated by disruption, as in the parental YS1646 strain.

A. Deletion of purl Gene Fragments and Transposon-Associated Insertion Sequence Elements

A first region, located between the yffB and purN genes, and containing: 1) a 1,209 bp transposon insertion sequence element, annotated as DC51_2586 in the sequence obtained from Broadway et al. (2014) (see, GenBank Accession numbers CP007804 and CP008745); and 2) 740 bp of the remaining 891 bp purl gene fragment (referred to herein as the large purl gene fragment), was targeted for deletion, using modifications of the method of Datsenko and Wanner (see, Proc. Natl. Acad. Sci. U.S.A. 97:6640-6645 (2000)). A small 151 bp portion of the 891 bp (large) purl gene fragment was left intact to avoid affecting the adjacent, downstream gene, purN. A plasmid, pSL0165, containing 284 and 262 bps of homology to the left hand and right hand regions, respectively, of the DC51_2586 insertion sequence element and the purl gene fragment, was transformed into DH5 -alpha competent cells (Thermo Fisher Scientific). A kanamycin gene cassette flanked by loxP sites was cloned into this plasmid, and the resulting vector was designated pSL0174. The DC51_2586 insertion sequence element and the large purl gene fragment knockout cassette then was PCR amplified using primers purm-1 and purm-2 (SEQ ID NOs: 419 and 420, respectively; see Table 2 below), gel purified, and introduced into strain YS I646At/.szZ ^/ AFLG/ NpagPI SansBI Scsg ) carrying the temperature sensitive lambda red recombination plasmid pKD46, by electroporation. The kanamycin resistance gene then was cured by Cre-mediated recombination, as described above, and the temperature-sensitive plasmids were cured by growth at non-permissive temperature. The DC51_2586 insertion sequence element and large purl gene fragment knockout sequences were confirmed by PCR using primers purm-3 and purm-4 (SEQ ID NOs: 421 and 422, respectively; see Table 2), and verified by DNA sequencing. The resulting mutant derivative of parental strain YS1646Aa«Z/AFLG/A/ifl'gP/A<»7£8/AcsgD was designated YS 1646 Aasz// SYLGISpagPISansBIScsgDI Spurl (large clean) .

A second region, located between the acrD and DC51_2568 genes, and containing: 1) a 1,209 bp transposon insertion sequence element, annotated as DC51_2566 in the sequence obtained from Broadway etal. (2014) (see, GenBank Accession numbers CP007804 and CP008745); and 2) the remaining 231 bp purl gene fragment (referred to herein as the small purl gene fragment), was targeted for deletion, using modifications of the method of Datsenko and Wanner (see, Proc. Natl. Acad. Sci. U.S.A. 97:6640-6645 (2000)). A plasmid, pSL0210, containing 241 and 265 bps of homology to the left hand and right hand regions, respectively, of the DC51_2566 insertion sequence element and the small purl gene fragment, was transformed into DH5-alpha competent cells (Thermo Fisher Scientific). A kanamycin gene cassette flanked by loxP sites was cloned into this plasmid, and the resulting vector was designated pSL0212. The DC51_2566 insertion sequence element and the small purl gene fragment knockout cassette then was PCR amplified using primers acrd-1 and purm-5 (SEQ ID NOs: 425 and 423, respectively; see, Table 2), gel purified, and introduced into strain YS 1646Ac/.s't/ ^/ AFLG/ SpagP! SansB! ScsgDI Aj>wr/(iarge clean) carrying the temperature sensitive lambda red recombination plasmid pKD46, by electroporation. The kanamycin resistance gene then was cured by Cre-mediated recombination, as described above, and the temperature-sensitive plasmids were cured by growth at non- permissive temperature. The DC51_2566 insertion sequence element and small purl gene fragment knockout sequences were confirmed by PCR using primers purm-6 and acrd-3 (SEQ ID NOs: 424 and 426, respectively; see Table 2), and verified by DNA sequencing. The resulting mutant derivative of parental strain YS IbdbAc/.s't// AFLG/ SpagPt SansB! ScsgDI A/i?//7ii _ai g _{= cic:ill)} was designated as YS 16A6Sasdl NPVGINpagPISansBIScsgDI Spur I (f _U n _c iean), or alternatively, YS 1646 Sasd! NVLG!SpagPiSansBIScsgl)N -Spur I. Table 2. Primer Sequence Information

Fwd= forward; Rev = reverse; KO = knockout

Plasmid Information B. Strain YS1646AasJ/AFLG/ApagP/AansB/Acs^Z)/F-Apwr/Displays

Enhanced Injection Stock Cell Viability

To evaluate the effects of the deletions of the remaining purl gene sequence fragments and the transposon-associated insertion sequence elements on the in vitro fitness of the bacteria, the cell viability of strains of YS1646Atz5t//AFLG/AptzgP/A«n5 _J 8/Ac5gD, with and without the full purl gene deletion (F-Apz/rT), and carrying the same plasmids, was compared. The cell viability of strains of YS1646Aasa/AFLG/A/?agF/A6OTsI?/AcsgD/F- kpurl, containing either plasmid ADN-657 (pATI1.76 CMV muIL- 15Ra-IL- 15sc_T2A_huSTING N154S/R284G tazCTT HPRE bGHpA), or plasmid ADN-750 (pATI2.1 CMV VCIP huIL-15Ra-IL-l 5 sc_T2A_huSTINGN154S/R284G tazCTT HPRE bGHpA), and the cell viability of strains of YS1646Aa.sc/ ^/ AFLG/ \pagPI kansBt kcsgD, expressing the same plasmids (ADN-657 and ADN-750), was evaluated by directly comparing viable CFUs after culture processing for frozen injection stocks. Plasmid ADN-657 encodes murine IL-15Ra- IL-15sc and a modified human STING chimera, with a replacement of the CTT of human STING with the CTT of Tasmanian devil STING, and the GOF mutations N154S/R284G (huSTINGN154S/R284G tazCTT). The two payloads are expressed under the control of a CMV promoter, from a bicistronic construct comprising a T2A peptide. The construct also includes a Hepatitis B virus Posttranscriptional Regulatory Element (HPRE), and a bovine growth hormone polyadenylation signal sequence (bGHpA). Plasmid ADN-750 encodes human IL-15Rci-IL-15sc and the same modified human STING chimera, where expression of the two payloads is under control of a CMV promoter, and the single promoter system is achieved using an endogenous human IRES (Internal Ribosome Entry Site), known as Vascular Endothelial Growth Factor and Type 1 Collagen Inducible Protein (VCIP; SEQ ID NO:434), placed upstream of the nucleic acid encoding both payloads, as well as a T2A peptide sequence, placed between the two ORFs.

For this experiment, 100 pl of overnight stationary phase cultures, of equivalent ODeoonm optical density (OD) values, were used to inoculate 25 ml of 4XYT media in 250 ml baffled shaker flasks with vented caps, and the cultures were incubated at 37 °C with shaking (at 225 RPM) for approximately 6 hours. Cultures were harvested at stationary phase at equivalent ODeoonm values, washed twice, adjusted to ODeoonm = 2, ali quoted, and frozen at -80 °C. The following day, two aliquots of each strain were thawed, the ODeoonm values were measured, and the strains were plated on agar plates to determine the titer and viability. % viability was determined from the ratio of ODeoonm to CFUs/ml.

Strain YS 1646 \asdi AFLG/ ^pagPl^ansBI\csgDIV - A/w/-(A DN-657) injection stock was determined to be 77% viable, while strain YS1646Aa%// AFLG/ SpagP! kansB! Ac.sg/J-(ADN-657) injection stock was determined to be 62% viable. Strain YS1646Aas£//AFLG/A/?agP/AansB/AcsgD/F- A/w/-(ADN-750) injection stock was determined to be 72% viable, while strain YS1646Aa%// AFLG/ SpagPt kansB! Ac.sg/J-(ADN-750) injection stock was determined to be 63% viable.

These data demonstrate a similar or enhanced fitness profile for strain YS1646Afl'5t//AFLG/Apfl'g7 ^? /Arz«55/Ac5gD/F-A/2Mr7 compared to the parental strain, YS 1646 Aasd/ AFLG/ L^pctgP / kansBI KcsgD . Strain YS1646Aa«Z/AFLG/A/ragP/A<»7s7?/AcsgD/F-A/>M77 displayed improved viability following frozen injection stock preparation, indicating that the genomic deletions not only diminish a potential genetic instability issue, but also provide a metabolic benefit that is manifested as increased cell viability.

C. Strain YS1646Au.$<// AFLG/ SpagP/ \anslil Ac.s^D/F-A/u/r/ Displays Similar Growth Characteristics in Broth Media Compared to the YS 1646 Aasv// AFLG/ \pagPI kansBI \csgl) Parental Strain

To evaluate the effects of the deletions of the remaining purl gene sequence fragments and the transposon-associated insertion sequence elements on the in vitro fitness of the bacterial strains, broth medium growth was compared between strains of YS IbdbAc/.s't// AFLG/ kpagPI kansBI \csgl)'V-\purI^ expressing either plasmid ADN- 657 (pATI1.76 CMV muIL-15Ra-IL-15sc_T2A_huSTINGN154S/R284G tazCTT HPRE bGHpA) or plasmid ADN-750 (pATI2.1 CMV VCIP huIL- l 5Ra-IL- 15sc_T2A_huSTINGN154S/R284G tazCTT HPRE bGHpA), and strains of YS 1646A<7.st/ ^/ AFLG/ \pagPI t^ansBI \csgP), expressing the same plasmids (ADN-657 and ADN-750). Frozen injection stocks were thawed at room temperature, and normalized to 1X10 ⁷ CFU/mL by diluting in PBS. 10 pL of the normalized samples were used to inoculate 300 pL of LB media (lx 10 ⁵ CFU/well) in a clear, flat- bottomed 96-well plate in technical quadruplicate. The plate was incubated with shaking at 37 °C, and the ODgoonm values were monitored in 15 minute intervals over the course of 16 hours. ODgoonm values were plotted to construct growth curves, and the slope of the log phase of growth was calculated and used to determine the doubling time for each strain.

Each of strains YS 1646 kasd! AFLG/ \pagPi kctnsB! \csgl)!V-\purI-( ADN- 657), YS1646 Aasd/ AFLG/ \pag!P \ansBi &csgI)-( ADN-657), YS 1646 \asd! AF LG/ ^pagPi^ansBI\csgl)IY-\purI-{\V)^-J 50) and YS 1646A<zst/ ^/ AFLG/ kpagPI \ansBI Ac.sg/J-(ADN-750) generated comparable growth profiles, and reached similar cell densities at stationary phase. Strain YS I646Aa.sa' ^/ AFLG/ kpagPI kansB! AcsgD/F -A/w/7-(ADN-657) had a doubling time of 77 minutes, strain YS1646Aasd/i\FLG/i\pagP/^ansB/ Ac.sg7J-(ADN-657) had a doubling time of 62 minutes, strain YS1646Aast//AFLG/ApagP/Aansfi/Ac.sgD/F- A/w/-(ADN-750) had a doubling time of 72 minutes, and strain YS I646Aa.st/ ^/ AFLG/ t^pagPi kansBI Ac.sg/J-(ADN-750) had a doubling time of 63 minutes. These data demonstrate a similar fitness profile for strain YS1646Aas4//AFLG/A/?agP/Aa«y8/AcsgD/F-A/>M77 compared to the parental strain, YS 1646 \asd! AFLG/ \pagPI\ansBlkcsgD. Strain

YS 1646Aa.se/ AFLG/ \pagP/ kansBt AcsgD/F -kpurl displayed slightly increased doubling time in broth medium, but generated similar growth curve profiles and stationary phase cell densities, compared to the parental strain.

EXAMPLE 20

Genetically Modified Strains Reduce Inflammation in Whole Human Blood and in Primary Human Macrophages

To evaluate the inflammatory profiles resulting from the administration of different variants of the immunostimulatory bacterial strains, strains ATCC # 14028 (wild-type (WT) S. typhimurium), YS1646, YS I646Aa.se/ YS I 646Aa.sc//AFLG, YS 1646 kasd! AFLG/ \pagP , YS 1646 Aa.sc// AFLG/ tSpagPI kansB! \csgl), YS 1646Aa.s'c/ ^/ AFLG/ bpagPI&ansBI KcsgDft-hpurl, and /s. coli strain NEB 5-alpha, each carrying plasmids (CMV NanoLuc®) that encode the luciferase NanoLuciferase® (Promega), were incubated with human blood for 2 hours at 37 °C. The incubations were in performed in a 96-well format, with 200 pl of blood and 5xl0 ³ CFUs of bacteria. At 2 hours post-infection, the blood was centrifuged at 300 relative centrifugal force (ref) for 5 minutes, and serum was isolated for cytokine profile analysis by human antiviral cytokine bead array analysis (BioLegend), in accordance with the manufacturer’s instructions.

Analysis of the levels of cytokines released from the human blood, following incubation with the various strains, showed that incubation with strain YS1646 resulted in reduced the levels of the pro-inflammatory cytokines IL-6 and TNF-a, compared to incubation with WT strain ATCC 14028. Additional genomic modifications further reduced the levels of pro-inflammatory cytokines released from the human blood.

STDEV = standard deviation

To determine the how the genomic modifications in bacteria, such as S. typhimurium influence inflammation during infection of primary human myeloid cells, bactofection of human M2 macrophages was performed. Primary human monocytes were differentiated in ImmunoCult™- SF Macrophage Medium (StemCell Technologies), containing 100 ng/ml human macrophage colony-stimulating factor (M-CSF), for 5 days. On day 6, cells were supplemented with additional medium containing 150 ng/ml M-CSF and 60 ng/ml huIL-4 for 48 hours, to generate M2 macrophages. The cells then were infected with stationary phase bacterial strains (strains described above), which were grown overnight at 37 °C in 4XYT medium, at an MOI of 10. The cells were inoculated with bacteria, and centrifuged for 5 minutes at 500 ref, followed by an incubation at 37 °C for 1 hour. The cells then were washed twice with DPBS, and then incubated in fresh ImmunoCult™- SF Macrophage Medium with 100 pg/ml gentamicin. Supernatants were harvested at 0, 2, 6, 24, and 48 hours post-infection. The cytokine measurements were performed by MESO scale analysis, according to the manufacturer’s instructions.

The results, which are summarized in the table below, show that, at 24 hours post-infection, macrophages infected with strains YS1646, YS I646Aczsc/, and YS I646AC/.SC//AFLG released the highest levels of IL-6, with WT bacteria (strain ATCC 14028) inducing less IL-6. Successive genomic modifications resulting in decreased levels of IL-6 secretion, with strains YS I646Acz.sc// AFLG/ pagPI csgl) and YS I646Ac7.sc// AFLG/ pagPI ansB/ csgl) inducing the lowest levels of IL-6. Similar trends were observed with TNF-a.

EXAMPLE 21 csgD-Deleted Strains Induce Vascular Leakage in Human Vascular Endothelial Cells

The ability to promote vascular leakage is an advantageous feature for facilitating bacterial entry into the tumor vasculature, and for promoting greater tumor colonization. It is shown in this example that kcsgD strains possess this feature.

Parental strain YS1646 (VNP20009) was compared to the derivative strains YS1646Aasz//AFLG/A/?^g/’ and YS 1646Amt//AFLG/A/%/g/VAr//7.s7//Ac'.s _i g/L for the ability to activate innate immune sensing in human umbilical vein endothelial cells (HUVECs), and for the ability to stimulate a subsequent increase in endothelial monolayer permeability. This was assessed using an In Vitro Vascular Permeability Assay kit (Millipore, Catalog No. ECM642). Strains of YS1646A«5t7/AFLG/Aj>«gP and Y S I 646 Ac/.w.// AFLG/ kpagPI KansB! Lscsglp each containing a plasmid encoding secreted NanoLuciferase® (EF-la secNanoLuc®), were used. Human umbilical vein endothelial cells (HUVECs) were seeded onto semi-permeable collagen-coated membrane inserts, in a 96-well plate chamber, according to the manufacturer’s instructions, at a concentration of 5xlO ⁵ cells/well. Confluency was monitored daily, until endothelial monolayer formation was confirmed, and determined to efficiently occlude the membrane pores (96 hours post-seeding).

The bacterial strains were grown in 3 mL of 4XYT media (TEKNOVA, Catalog No. 2Y1085), with shaking at 37 °C in vented cap 50 mL conical vials, overnight to stationary phase, and were prepared the following day by pelleting and re-suspending in PBS. Bacterial samples were normalized by ODeoonm to a concentration of 2.5X 10 ⁸ CFU/mL, and were added to the HUVEC insert wells in 100 pL volume, to achieve an MOI of 50. The plate was spun at 500 ref for 5 minutes to synchronize engagement of CFUs with HUVECs, and then the plate was incubated for 1 hour at 37 °C. Gentamicin was then added to the insert wells to a final concentration of 200 pg/mL. At 24 hours post-bactofection, medium in the inserts was collected (and saved for cytokine analysis), and 75 pL of Fluorescein isothiocyanate-Dextran (FITC-Dextran; at a 1 :40 dilution) was added, and the mixture was incubated in the dark, at room temperature, for 20 minutes. Medium from the receiver tray was collected and analyzed on a spectrophotometer at an excitation wavelength of 485 nm and an absorbance wavelength of 535 nm, and 1 :40 FITC- Dextran solution was run directly as a positive control.

At 24 hours post-bactofection, and following incubation with FITC-Dextran solution, wells that were not treated with engineered, attenuated 5. typhimurium strains (i.e., medium alone) generated very low fluorescence values (35), indicating confluency and little disruption of permeability. The parental YS1646-treated wells permitted the greatest amount of FITC-Dextran to pass through the membrane (fluorescence = 155), indicating the strongest innate immune stimulation and increase in vascular permeability, while the engineered, attenuated S. typhimurium strains YS l 646Ac/.st/ ^/ AFLG/A/?c/ _< t, ^r / ^> and YS I 646AC/.S6// AFLG/ \pagPI AansBI kcsgli) permitted less FITC-Dextran to pass through (fluorescence = 84 and 91, respectively), yet still maintained the ability to promote vascular permeability.

To test whether this effect was dependent on cytokines released from the HUVECs, and in particular, IL-6, a human antiviral cytokine bead array analysis (BioLegend) was performed on HUVEC supernatants at 6 hours post-infection, in accordance with the manufacturer’s instructions. HUVECs infected with the parental YS1646 strain exhibited very high IL-6 levels (1769.2 pg/mL), compared to the uninfected control (27.19 pg/mL). The YS1646 kasd! AFLG/ tspagP strain induced lower levels of IL-6 (543.3 pg/mL), and the lowest amount of IL-6 was observed with the YS1646Aa%//AFLG/A/?fl'gP/Arzns.8/AcsgD strain (272.8 pg/mL). Therefore, despite producing lower levels of pro-inflammatory cytokines that induce vascular leakage, the YS1646Aa%//AFLG/A/?^gP/A£OTxS/AcsgD strain demonstrated higher levels of leakage than the YS1646 ^asd! AFLG/ kpagP strain. Thus, the ability to promote vascular leakage is an important feature for facilitating bacterial entry into the tumor vasculature, and for promoting greater tumor colonization. Strains described and provided herein possess this feature.

EXAMPLE 22

STING Gain-Of-Function Variants Demonstrate Increased CXCL10 (IP-10) to IL-6 Ratios and Increased IFN-P to IL-6 Ratios in Primary Human M2 Macrophages

To determine the effects of various STING gain-of-function (GOF) mutants (including chimeric STING proteins with GOF mutations), on the downstream cytokine signaling in human M2 macrophages, the GOF mutants were cloned into the pATI-1.75 (also referred to as pATI1.75) vector, described in Example 8 above, which then was transfected into human M2 macrophages for expression, and the cell supernatants were assayed for secreted (expressed) cytokines.

Frozen human monocytes, isolated from healthy human donors, were thawed in complete medium (RPMI-1640 + 10% FBS), and washed by centrifugation for 10 minutes at 600 x g, at room temperature. The monocytes were resuspended in ImmunoCult™- SF Macrophage Medium (StemCell Technologies) containing 100 ng/mL of human M-CSF + 20 ng/mL of human IL-4 + 20 ng/mL of human IL- 10. Monocytes (8e5 to le6 cells per well) were then seeded in a 24-well plate, with a final volume of 750 uL. Three days later, 750 pL of ImmunoCult™- SF Macrophage Medium (StemCell Technologies), containing 100 ng/mL human M-CSF + 20 ng/mL human IL-4 + 20 ng/mL human IL- 10, was added to each well, and the plate was incubated for four more days. On day seven, the cells were transfected using Viromer® RED mRNA and plasmid transfection reagent (Lipocalyx GmbH), according to the manufacturer’s instructions. 500 ng of plasmid DNA from a panel of STING GOF mutants, as well as untransfected control, were diluted in the provided buffer, and mixed with the Viromer® RED transfection reagent, and incubated at room temperature for 15 minutes to allow the DNA/Viromer® RED complexes to form. The DNA/Viromer® RED complexes were then slowly added to each well of the 24-well plate (in triplicate), and the plate was incubated at 37 °C in a 5% CO ₂ incubator. Supernatants were harvested after 48 hours of incubation, and were assayed for human CXCL10 (IP- 10) and IL-6 using a flow cytometry -based human anti-viral cytokine bead array (BioLegend), according to the manufacturer’s protocol. The average of three measurements was calculated, and the ratio of IP-10 to IL-6 was calculated by dividing the IP- 10 concentration by the IL-6 concentration.

The results, which are summarized in the table below, show that the STING variant that resulted in the highest ratio of IP- 10 to IL-6 was the huSTING N154S/R284G tazCTT variant (i.e., a chimeric STING protein, containing a modified human STING with a replacement of the CTT with the CTT of Tasmanian devil STING, and the GOF mutations N154S/R284G). The STING variant, containing a replacement of the CTT of human STING with the CTT of Tasmanian devil STING (huSTING tazCTT), displayed higher ratios of IP-10 to IL-6 than the corresponding fully human STING variants with the same mutations, thus, demonstrating an improved anti-tumor response.

Ratio of IP-10 to IL-6 Protein Expression Following Transfection of M2 Macrophages with plasmids encoding STING Variants

The ratio of IFN- to IL-6 gene expression in human M2 macrophages, following the expression of various STING variants, was assessed. Frozen human monocytes, isolated from healthy human donors, were thawed in complete medium (RPMI-1640 + 10% FBS), and washed by centrifugation for 10 minutes at 600 x g, at room temperature. The monocytes were resuspended in RPMI-1640 + IX non- essential amino acids (NEAA) + 5% human AB serum, containing 200 ng/mL human M-CSF + 20 ng/mL human IL-4. Monocytes (8e5 to le6 cells per well) were then seeded in a 24-well plate, with a final volume of 750 pL. Three days later, 750 pL of RPMI with 5% human AB serum + NEAA, containing 200 ng/mL human M-CSF + 20 ng/mL human IL-4, was added to each well, and the plate was incubated for four more days. On day seven, the cells were transfected using Viromer® RED transfection reagent, according to the manufacturer’s instructions. 500 ng of plasmid DNA from a panel of STING GOF mutants, including the modified chimeric STING proteins, as well as untransfected control, were diluted in the provided buffer, and mixed with Viromer® RED transfection reagent, and incubated at room temperature for 15 minutes to allow the DNA/Viromer® RED complexes to form. As a positive control, the STING agonist 3’5’ RpRp c-di-AMP (InvivoGen), an analog of the clinical compound ADU-S100, was added to the cells at a concentration of 10 pg/mL. The DNA/Viromer® RED complexes then slowly were added to each well of the 24- well plate (in triplicate), and the plate was incubated for 48 hours at 37 °C, in a CO ₂ incubator.

The cells were harvested for qPCR 48 hours post-transfection, and were lysed with 350 pL Buffer RLT lysis buffer with beta-mercaptoethanol (Qiagen). RNA extraction was performed using the Qiagen RNeasy® Plus Mini Kit, with the following modifications. A genomic DNA elimination step, using an RNase-Free DNase kit (Qiagen), was included in the kit to remove genomic DNA from the total RNA. Total RNA concentration was measured using a NanoDrop™ One ^c UV-Vis Spectrophotometer (Thermo Fisher Scientific). The purity of each sample also was assessed from the A260/A230 absorption ratio. RNA was stored at -80 °C without freeze-thawing, until reverse-transcription was performed.

Synthesis of cDNA was performed from 0.5-1 pg of template RNA, using a CFX96™ Real-Time System (Bio-Rad) and iScript™ Reverse Transcription Supermix for RT-qPCR (Bio-Rad) in a 20 pL reaction, according to the manufacturer’s instructions. qPCR was performed with a CFX96™ Real-Time System (Bio-Rad). Primers for human IFN 1 (huIFNpi; Assay ID: qHsaCEP0054112; Bio-Rad) and for huIL-6 (Assay ID: qHsaCEP0051939; Bio-Rad) were used for the qPCR. The qPCR reaction (20 pL) was conducted per protocol, using either the SsoAdvanced™ Universal SYBR® Green Supermix, or the iQ™ Multiplex Powermix (Bio-Rad). The standard thermocycling program on the Bio-Rad CFX96™ Real-Time System comprised a 95 °C denaturation step for 150 seconds, followed by 39 cycles of 95 °C for 15 seconds, and 60 °C for 55 seconds. Quantification of the target mRNA was normalized using actin reference mRNA (Bio-Rad, Assay ID: qHsaCEP0036280). AC _q was calculated as the difference between target (huIFNP or huIL-6) and reference (actin) gene. AAC _q was obtained by normalizing the AC _q values of the treatments (transfections), to the AC _q values of the non-treatment (i.e., untransfected) controls. The ratios of the IFNP AAC _q to the IL-6 AAC _q are shown in the table below. The results, which are summarized in the table below, show that, of all of the STING GOF mutants screened, the huSTINGN154S/R284GtazCTT variant results in the highest ratio of immunostimulatory IFN-P to pro-inflammatory IL-6 expression. Additionally, the chimeric STING constructs, containing a replacement of the CTT of human STING with the CTT of Tasmanian devil STING (e.g. , huSTING tazCTT), generally induced higher ratios of IFN-P to IL-6 expression than the corresponding fully human STING constructs comprising the same GOF mutation(s).

Increased expression of IFN-P indicates increased IRF3/type I IFN signaling, which is immunostimulatory and beneficial, while decreased IL-6 expression is indicative of reduced NF-KB signaling, which is pro-inflammatory and does not contribute to an anti -turn or response. A higher ratio of IFN-P to IL-6 expression is indicative of an increased anti-tumor/anti-viral type response, and a decreased pro- inflammatory response. Thus, replacement of the CTT, as well as the gain-of-function mutations, in a STING protein, increases the anti-tumor activity of the STING protein, and, hence, the immunostimulatory bacterium.

Ratio of IFN-P to IL-6 Gene Expression Following Transfection of M2

Macrophages with STING Variants

EXAMPLE 23

Design of IL-15 Receptor-a and IL-15 Single-Chain Fusion Proteins

Fusion proteins, containing human proteins or mouse proteins, were prepared. The mouse proteins are for use in mouse models; the human proteins are for encoding in the immunostimulatory bacteria to be used as human therapeutics.

A human IL-15 Receptor-a (IL-15Ra) fused to a human IL-15 single-chain (sc) was designed, as follows. Amino acid residues 1-108 of human IL-15Ra (SEQ ID NO:401), which correspond to the leader sequence and sushi domain of IL-15-Ra (plus an additional 13 amino acid residues of IL-15Ra; see, e.g., Bouchaud et al. (2008) J. Mol. Biol. 382(1): 1-12) were added in-frame with a Gly-Ser linker that has four repeats of the sequence Gly-Gly-Gly-Gly-Ser (z.e., (GGGGS)4). The polypeptide sequence corresponding to fully mature human IL- 15 sc, without the leader sequence or propeptide, and corresponding to amino acid residues 48-162 of SEQ ID NO:403, was added following the linker. The sequence of the resulting human IL-15Ra-IL- 15sc fusion protein (SEQ ID NO:404) is as follows: MAPRRARGCRTLGLPALLLLLLLRPPATRGITCPPPMSVEHADIWVKSYSLYS RERYICNSGFKRKAGTSSLTECVLNKATNVAHWTTPSLKCIRDPALVHQRPAP PSGGGGSGGGGSGGGGSGGGGSANWVNVISDLKKIEDLIOSMHIDATLYTE SDVHPSCKVTAMKCFLLELOVISLESGDASIHDTVENLIILANNSLSSNGNVTE SGCKECEELEEKNIKEFLOSFVHIVOMFINTS. where the residues corresponding to residues 1-108 of human IL-15Ra are underlined; the Gly-Ser linker is in bold letters; and the residues corresponding to residues 48-162 of human IL-15 are double underlined.

Similarly, a murine IL-15Ra-IL-15sc fusion protein (SEQ ID NO:407) was prepared by fusion of a portion of the murine IL- 15 Receptor-a (IL-15Ra) protein to a portion of the murine IL- 15 single-chain (sc) protein. Amino acid residues 1-132 of murine IL-15Ra (SEQ ID NO:405), which include the leader sequence and sushi domain of IL-15-Ra, were added in-frame with a (GGGGSp linker. The polypeptide sequence corresponding to fully mature murine IL-15sc, without the leader sequence or propeptide, and corresponding to amino acid residues 49-162 of SEQ ID NO:406, was added following the linker. The sequence of the resulting murine IL-15Rot-IL- 15sc fusion protein (SEQ ID NO:407), which was used for mouse model experiments, is as follows:

MASPQLRGYGVQAIPVLLLLLLLLLLPLRVTPGTTCPPPVSIEHADIRVKNYSV NSRERYVCNSGFKRKAGTSTLIECVINKNTNVAHWTTPSLKCIRDPSLAHYSP VPTVVTPKVTSQPESPSPSAKEPEAGGGGSGGGGSGGGGSGGGGSNWIDVR YDLEKIESLIOSIHIDTTLYTDSDFHPSCKVTAMNCFLLELOVILHEYSNMTLN ETVRNVLYLANSTLSSNKNVAESGCKECEELEEKTFTEFLOSFIRIVOMFINTS. where the residues corresponding to residues 1-132 of murine IL-15Ra are underlined; the Gly-Ser linker is in bold letters; and the residues corresponding to residues 49-162 of murine IL- 15 are double underlined.

EXAMPLE 24 IL-15Ra-IL-15sc Induces Curative Effects and Protection from Tumor ReChallenge in a Mouse Model of Colorectal Carcinoma

This example demonstrates that immunostimulatory bacteria that encode the IL-15Ra-IL15sc fusion protein (also referred to herein as IL-15/IL-15R alpha chain complex, IL-15/IL-15Ra chain complex, and IL-15/IL-15Ra) induce anti-tumor efficacy as a monotherapy. To demonstrate this, S. typhimurium strains, containing plasmids encoding murine IL-15Ra-IL15sc (muIL-15Ra-IL15sc), were prepared. The YS I646Ac/.w.Z ^/ AFLG/ BpagP! kansBI BcsgD strain, containing the plasmid encoding muIL- 15Ra-IL 15 sc (/. e. , YS 1646 kasd! AFLG/ kpagPI kcmsB/ AcsgD-m u IL- 15Ra- IL15sc; mouse IL-15Ra-IL15sc was used for the experiments performed in the mouse model), was compared to PBS control, for safety and efficacy in the subcutaneous (SC) flank MC38 colorectal adenocarcinoma model. For this study, 6-8 week-old female C57BL/6 mice (5 mice per group) were inoculated SC in the right flank with MC38 cells (5xlO ⁵ cells in 100 pL PBS). Mice bearing established flank tumors were intravenously (IV) injected on day 8 with 2xl0 ⁷ CFUs of the YS1646Aa%//AFLG/A/ragP/Aa'ns/?/AcsgD-muIL-15Ra-IL15sc strain, or with PBS vehicle control. Tumor measurements and body weights were recorded twice weekly.

The results showed that the YS1646Aasd/BFLG/ApagP/&msB/AcsgD-muIL- 15Ra-IL15sc strain demonstrated 50% cures (4/8 vs. 0/8 for PBS,/? = 0.005, day 21), after a single IV injection. At 66 days post-tumor implantation (day 57 post-IV dosing), cured mice (N=4) were re-implanted, on the opposite flank, with 5xlO ⁵ MC38 cells, and tumor growth was compared to naive, age-matched mice (N=5). By day 30 post re-implantation, all mice in the group cured with strain

YS 1 (Ab&asdl AFLG/ kpagP/ kansB/ Ac.sg/J-muIL- 15Ra-IL 15 sc remained tumor-free, whereas all mice in the naive group had reached maximum tumor volume. These data demonstrate the potent and curative effects of delivery of the IL-15Ra-IL15sc fusion protein via an immunostimulatory bacterium, such as strain YS I646Ac/.st/AFLG/A/?c/g/7A<:7//.s///Ac ^, .sg/J-muIL- l 5Ra-ILI 5sc. and demonstrate the induction of durable protective immune memory in a model of colorectal carcinoma.

EXAMPLE 25

Anti-CTLA-4 scFv-Fc Demonstrates Superior Blockade of the CD80/CTLA-4 and CD86/CTLA-4 Interactions Compared to Anti-CTLA-4 scFv

An scFv-Fc specific for human CTLA-4 (see, SEQ ID NOs:427 and 428 for the nucleic acid and protein sequences, respectively) was designed using the amino acid sequence of ipilimumab. Ipilimumab (YERVOY®) is a fully human IgGlK monoclonal antibody that specifically binds human CTLA-4 (see, e.g., the antibody designated 10D1 in U.S. Patent Publication No. 2002/0086014 and in U.S. Patent No. 6,984,720), blocking CTLA-4’ s immune inhibiting interaction with CD80 (also known as B7.1 or B7-1) and CD86 (also known as B7.2 or B7-2).

To generate the ipilimumab scFv antibody fragment (see, SEQ ID NO:429), the variable light chain (VL) and variable heavy chain (VH) of ipilimumab were linked with a 20 amino acid long glycine-serine (GS) linker ((GGGGS)4). To generate the scFv-Fc antibody fragment (see, SEQ ID NO:428), the variable heavy chain of the ipilimumab scFv was linked to a human IgGl Fc, containing a mutation of the free cysteine in the hinge region to a serine (at position 272 in SEQ ID NO:428). The leader sequence (METPAQLLFLLLLWLPDTTG; corresponding to residues 1-20 in SEQ ID NO:428) was derived from the sequence of the human immunoglobulin kappa variable 3-20 (IGKV3-20) protein. The sequence was codon optimized using the GenScrip GenSmart™ Codon Optimization tool.

The neutralizing ability of the anti-CTLA-4 scFv-Fc was compared to that of the anti-CTLA-4 scFv (lacking the human IgGl Fc portion), using competitive ELISAs to measure the ability of each of the antibody fragments to block the interactions between CTLA-4 and its ligands, CD80 and CD86. HEK293T cells were transfected with 3 micrograms of DNA encoding the anti-CTLA-4 scFv-Fc or the anti-CTLA-4 scFv antibody fragment constructs, using the FuGENE® transfection reagent (Promega), at the proper reagentDNA ratios. Forty-eight hours posttransfection, HEK293T cell-free culture supernatants were harvested, filtered, and used in a competitive ELISA to assess the blockade activity of the anti-CTLA-4 antibody fragments.

For the competitive ELISAs, mouse CD80 or CD86 recombinant proteins (R&D Systems) were coated overnight at 4 °C on a high protein-binding 96-well plate, at a concentration of 100 ng/ml. The wells were then washed one time with PBS 0.05% Tween-20, and the wells were blocked with ELISA blocking buffer for 1 hour at room temperature. The wells were then washed one time with PBS 0.05% Tween- 20. HEK293T cell culture supernatants, containing each of the anti-CTLA-4 antibody fragments, were mixed with 10 ng/ml of a recombinant murine CTLA-4-human IgGl Fc chimera (R&D Systems), and added to the wells and incubated for 2 hours at room temperature. The wells were then washed three times with PBS 0.05% Tween-20, and horseradish peroxidase (HRP)-conjugated anti-human IgGl antibody was added to the wells (Jackson ImmunoResearch), and incubated for one hour at room temperature. The wells were then washed three times with PBS 0.05% Tween-20, and detection reagent (3,3',5,5'-tetramethylbenzidine (TMB), Thermo Fisher Scientific) was added to the wells. The enzymatic reaction was stopped with sulfuric acid (BioLegend), and the optical densities were read at 450 nm.

The results of the competitive ELISAs are summarized in the table below. The anti-CTLA-4 scFv-Fc blocked the binding of CTLA-4 to CD86 by 75.5%, and blocked the binding of CTLA-4 to CD80 by 40.6%, whereas the anti-CTLA-4 scFv blocked the binding of CTLA-4 to CD86 by 32.5%, and blocked the binding of CTLA-4 to CD80 by 7%. A higher degree of CD86/CTLA-4 blocking activity (compared with CD80/CTLA-4 blocking activity) was observed with both antibody fragments, and a superior neutralizing activity was observed with the anti-CTLA-4 scFv-Fc, when compared to the anti-CTLA-4 scFv.

Competitive ELISA Results An anti-murine CTLA-4 scFv (SEQ ID NO:430) and anti-murine CTLA-4 scFv-Fc (SEQ ID NO:431), derived from the 9D9 clone, also were prepared, as discussed above for the corresponding human anti -CTLA-4 antibody fragments. The scFv contains an IgK leader mouse sequence, and the VL and VH domains from clone 9D9, linked via a (Gly4Ser)3 linker. The scFv-Fc also contains a mouse IgG2a Fc linked to the VH domain.

Included among the immunostimulatory bacteria provided herein, are those that encode, on the plasmid, anti-CTLA-4 antibodies and fragments thereof, including the anti-CTLA-4 scFv antibody fragments and the anti-CTLA-4 scFv-Fc antibody fragments, as provided herein, as well as combinations with nucleic acid molecules encoding other therapeutic products.

EXAMPLE 26

Optimized Expression Cassettes for High-Level, Multiplexed Expression of Encoded Therapeutic Products

Elements, known as Internal Ribosomal Entry Sites (IRES), can promote protein translation by enhancing ribosomal binding and stabilization of mRNA through mimicking the 5 ’mRNA cap. IRES elements from viruses and mammalian cells are well-known. Of these, an endogenous human IRES, known as Vascular Endothelial Growth Factor and Type 1 Collagen Inducible Protein (VCIP), previously was demonstrated to enhance expression from a bicistronic vector, encoding Firefly and Renilla luciferases, where the Renilla luciferase, which was the downstream gene (i.e., encoded after the VCIP IRES), was expressed in vitro and in vivo, without the need for a second promoter (see, e.g., Licursi et al. (2011) Gene Therapy 18(6):631- 636).

Plasmids containing combinations of human or mouse IL-15Ra-IL-15sc and huSTING N154S/R284GtazCTT, or human or mouse IL-15Ra-IL-15sc, huSTING N154S/R284G tazCTT, and an anti-CTLA-4 scFv-Fc, where the bicistronic and polycistronic constructs contain 2A peptides, were tested for expression of each encoded payload/product against single expresser controls. Additionally, combination constructs, containing the VCIP IRES, also were tested.

HEK293T STING Null Cells (293 -Dual™ Null Cells; InvivoGen) were used, which do not contain endogenous STING, and express secreted embryonic alkaline phosphatase (SEAP), placed under the control of the endogenous IFN-stimulated response element (ISRE) promoter, where the coding sequence of ISRE is replaced by the SEAP ORF using knock-in technology. STING activity can thus be assessed by monitoring type I IFN induced SEAP production. The 293-Dual™ Null cells also express Lucia™ luciferase, a secreted luciferase, placed under the control of the endogenous IFN-P promoter; the coding sequence of IFN- has been replaced by the Lucia™ luciferase ORF using knock-in technology. This allows for the assessment of STING activity by monitoring the expression of IFN-p. Using these cells, STING activity can be assessed by monitoring ISRE-induced SEAP production and/or IFN-P-dependent expression of Lucia™ luciferase. The two reporter proteins, SEAP and Lucia™ Luciferase, can be measured in the cell supernatant using standard assays and detection reagents, such as the QUANTLBlue™ and QUANTI-Luc™ detection reagents (InvivoGen), respectively.

The cells were seeded in 24-well plates coated with poly-L-lysine at 200,000 cells per well, and incubated overnight at 37 °C in a 5% CO2 incubator, to achieve 80% confluency. The following day, 300 ng of each plasmid DNA, and 40 ng of a CMV-GFP vector (i.e., a vector encoding green fluorescent protein under the control of a CMV promoter), were diluted in serum-free media and added to FuGENE® transfection reagent (Promega), at the proper reagent:DNA ratios, with untransfected wells serving as negative controls (in duplicates). Cell culture supernatants from each sample were collected 48 hours post-transfection.

The STING activity of the huSTINGN154S/R284G tazCTT variant was evaluated using the ISRE-SEAP and IFN-P-Lucia™ reporter systems. The type I interferon (IFN) activity (induced by STING) was assessed by monitoring type I IFN- stimulated SEAP production in the cell supernatants. 20 pL of cell culture supernatant was added to 180 pL of QUANTLBlue™ reagent (InvivoGen), which is used for measuring SEAP. Type I interferon activation was determined by measuring ISRE- induced SEAP activity on a SpectraMax® M3 Spectrophotometer (Molecular Devices), at an absorbance wavelength of 650 nm. The type I interferon (IFN) activity (induced by STING) also was assessed by monitoring the type I IFN-stimulated Lucia™ luciferase production in the cell supernatants. 20 pL of cell culture supernatant was added to 50 pL of QUANTI-Luc™ reagent (InvivoGen), which is used for measuring Lucia™ luciferase activity. Type I interferon activation was determined by measuring IFNP-induced Lucia™ luciferase activity on a SpectraMax® M3 Spectrophotometer (Molecular Devices), on the luminescence setting. Cell culture supernatants also were assessed for the expression of human or mouse IL-15Ra-IL-15sc (see, Example 23), and for the expression of human or mouse anti-CTLA-4 scFv-Fc (see, Example 25). For the muIL-15Ra-IL-15sc constructs, the murine IL-15Ra-IL-15sc ELISA (R&D) was used, per kit instructions. For the hu!L-15Ra-IL-15sc constructs, the human IL-15Ra-IL-15sc ELISA (R&D) was used, per kit instructions. Direct ELISAs, with human and mouse CTLA-4-Fc (R&D Systems), were performed on the cell culture supernatants of cells transfected with plasmids encoding anti-human CTLA-4 scFv-Fc and anti-murine CTLA-4 scFv- Fc, to measure the expression levels of these proteins.

GFP production was detected by flow cytometry, and was used to normalize the transfections to each other. 48 hours post-transfection, the cells were washed twice with PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes. The cells then were resuspended in PBS + 2% FBS with DAPI (dead/live stain). Flow cytometry data were acquired using the ACEA NovoCyte® flow cytometer (ACEA Biosciences, Inc ), and analyzed using the FlowJo™ software (Tree Star, Inc.).

As shown in the table below, all constructs containing the huSTING N154S/R284GtazCTT variant displayed ISRE-SEAP reporter activity and IFNp- Lucia™ luciferase reporter activity. The construct designated 2.1 CMV VCIP muIL- 15Ra-IL-15sc T2A huSTING N154S/R284G tazCTT, which contains the VCIP IRES and a T2A peptide for expression of the combination of muIL-15Ra-IL-15sc and huSTING N154S/R284GtazCTT, results in the highest expression levels of muIL- 15Ra-IL-15sc, when normalized by GFP co-transfection. Similarly, the construct designated 2.1 CMV VCIP huIL-15Ra-IL-15sc T2A huSTING N154S/R284G tazCTT, which contains the VCIP IRES and a T2A peptide for expression of the combination of huIL-15Ra-IL-15sc and huSTING N154S/R284G tazCTT, results in the highest expression levels of huIL-15Ra-IL-15sc, when normalized by GFP cotransfection. The construct designated 1.76 CMV mu anti-CTLA-4 scFv-Fc results in the highest level of expression of the murine anti-CTLA-4 scFv-Fc, and the construct designated 1.76 CMV hu anti-CTLA-4 scFv-Fc results in the highest level of expression of the human anti-CTLA-4 scFv-Fc, when normalized by GFP cotransfection.

Measurement of STING Activity, IL-15Ra-IL-15c Concentration, and Anti-CTLA-4 scFv-Fc Concentration, Normalized by GFP Co-Transfection, in HEK293T STING Null Cells

SD = Standard deviation *NA = Not applicable

EXAMPLE 27

Expression Plasmids Encoding the Combination of huIL-15Ra-IL-15sc + huSTING N154S/R284G tazCTT

To identify the plasmid(s), with the highest expression of huIL-15Ra-IL-15sc and huSTING N154S/R284G tazCTT, plasmids containing nucleic acid molecules encoding huIL-15Ra-IL-15sc and/or huSTING N154S/R284G tazCTT, with different 2A peptides e.g., T2A or P2A), and/or different post-transcriptional regulatory elements (e.g., HPRE or WPRE), and/or different poly(A) tails (e.g., bovine growth hormone poly(A) (bGHpA) or simian virus 40 poly(A) (SV40pA)), and/or containing the VCIP IRES, were cloned. Additionally, in some constructs, short peptide spacers, including RRKR and RAKR, and other spacers of different lengths, were encoded following the nucleic acid encoding the first payload (z.e., huIL-15Ra-IL-15sc), and before the nucleic acid encoding the 2A peptide. RRKR and RAKR are designed furin protease cleavage sites, and were placed in that position to facilitate proper processing of the 2A peptide sequences. The plasmids were first evaluated for expression and functionality by transfection in HEK293T STING Null Cells (ISG/KI-IFN p) cells (InvivoGen).

HEK293T STING Null Cells (293 -Dual™ Null Cells (ISG-SEAP/KI-[IFN- p]Lucia; InvivoGen), which do not contain endogenous STING, and express secreted embryonic alkaline phosphatase (SEAP), placed under the control of the endogenous IFN-stimulated response element (ISRE) promoter, where the coding sequence of ISRE is replaced by the SEAP ORF using knock-in technology, were used. As discussed above, the 293-Dual™ Null cells also express Lucia™ luciferase, a secreted luciferase, placed under the control of the endogenous IFN- promoter, where the coding sequence of IFN-p has been replaced by the Lucia™ luciferase ORF using knock-in technology. The cells were seeded in 24-well plates coated with poly-L- lysine, at 200,000 cells per well, and incubated overnight at 37 °C in a 5% CO2 incubator, to achieve 80% confluency. The following day, 300 ng of each plasmid DNA, and 40 ng of a CMV-GFP vector, were diluted in serum-free media and added to FuGENE® transfection reagent (Promega), at the proper reagent:DNA ratios, with untransfected wells serving as negative controls (in duplicates). Cell culture supernatants from each sample were collected 48 hours post-transfection for analysis.

The STING activity of the huSTINGN154S/R284G tazCTT modified STING protein was evaluated with the HEK293T STING Null (ISG-SEAP/KL[IFN-P]Lucia) reporter cell line (InvivoGen). Using these cells, the type I interferon (IFN) activity (that is induced by STING) is assessed by monitoring type I IFN-stimulated Lucia™ luciferase production in the cell supernatants. IFNP induction is measured with the IFNP-Lucia reporter. 20 pL of cell culture supernatant was added to 50 pL of QUANTI-Luc™ reagent (InvivoGen), which is used for measuring Lucia™ luciferase activity. Type I interferon activation was determined by measuring IFNP-induced Lucia™ luciferase activity on a SpectraMax® M3 Spectrophotometer (Molecular Devices), on the luminescence setting. Cell culture supernatants also were assessed for the expression of human IL-15Ra-IL-15sc by ELISA (R&D), per kit instructions.

GFP was detected by flow cytometry, and the expression levels of GFP (as measured by fluorescence) were used to normalize the transfections to one other. 48 hours post-transfection, the cells were washed twice with PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes. The cells then were resuspended in PBS + 2% FBS with DAPI (dead/live stain). Flow cytometry data were acquired using the ACEA NovoCyte® flow cytometer (ACEA Biosciences, Inc.), and analyzed using the FlowJo™ software (Tree Star, Inc ).

As shown in the two tables below, the plasmid construct designated 2.1 CMV VCIP huIL-15Ra-IL-15sc T2A huSTING N154S/R284GtazCTT HPRE bGHpA resulted in the highest normalized luminescence value from the IFNP- Lucia™ reporter (see, first table below), and also, the highest normalized concentration of expressed huZL-15Rot-IL-15sc, as determined by ELISA (see, second table below).

As shown in the first table below, the construct encoding only huIL- 15Ra-IL- 15sc (and no STING variant), as expected, does not exhibit STING-induced type I IFN activity. The construct encoding only the huSTINGN154S/R284G tazCTT exhibits a high level of STING-induced type I IFN activity. For the constructs encoding both payloads, the combination of a P2A peptide with WPRE results in higher levels of STING activity (/.<?., higher levels of expression of STING), than combinations of T2A and P2A with HPRE, or T2A with WPRE. Additionally, higher levels of STING activity were observed with constructs containing a bGH poly(A) tail, than an SV40 poly(A) tail. The addition of a short peptide spacer, such as RAKR or RRKR, slightly increases the level of STING activity, compared to the same constructs not containing these spacers. In certain constructs, such as those designated 2.1 CMV VCIP huIL-15Ra-IL-15sc T2A huSTINGN154S/R284G tazCTT HPRE bGHpA; 2.1 CMV VCIP huIL-15Ra-IL-15sc P2A huSTINGN154S/R284G tazCTT HPRE bGHpA; 2.1 CMV VCIP huIL-15Ra-IL-l 5sc T2A huSTING N154S/R284G tazCTT WPRE bGHpA; 2.1 CMV VCIP huIL-15Ra-IL-15sc T2A huSTING N154S/R284G tazCTT WPRE SV40pA; and 2.1 CMV VCIP huIL-15Rot- IL-15sc RAKR-T2A huSTING N154S/R284G tazCTT HPRE bGHpA, the addition of a VCIP IRES increases the levels of STING activity, compared to the same constructs not containing the VCIP IRES. As shown in the second table below, the expression levels of huIL-15Ra-IL- 15sc are higher from constructs containing a bGH poly(A) tail, than an SV40 poly(A) tail, and in general, constructs containing WPRE result in higher expression levels of huIL-15Ra-IL-15sc than the same constructs containing HPRE. The addition of an RAKR short peptide spacer, after the nucleic acid sequence encoding huIL-15Ra-IL- 15sc, in the construct encoding CMV huIL-15Ra-IL-15sc T2A huSTING N154S/R284GtazCTT HPRE bGHpA, increased the expression levels of huIL-15Ra- IL-15sc. In all constructs, except the one designated 2.1 CMV VCIP huIL-15Ra-IL- 15sc P2A huSTING N154S/R284G tazCTT HPRE SV40pA, the addition of a VCIP IRES (in addition to a 2A peptide) upstream of the construct, results in significantly increased expression of huIL-15Ra-IL-15sc. Replacement of the 2A peptide with a VCIP IRES, for example, in the construct designated 1.76 CMV huIL-15Ra-IL-15sc VCIP huSTING N154S/R284G tazCTT HPRE bGHpA, results in similar expression levels of huIL-15Ra-IL-15sc. Replacement of the 2A peptide with a VCIP IRES, and the addition of a longer spacer between the nucleic acids encoding huIL-15Ra-IL- 15sc and the VCIP IRES, for example, in the construct designated 1.76 CMV huIL- 15Ra-IL-15sc Longer spacer VCIP huSTING N154S/R284G tazCTT HPRE bGHpA, results in a significantly increased expression level of huIL-15Ra-IL-15sc. The spacer sequences in the constructs are nucleic acid sequences that are placed between the stop codon of the first ORF, and the VCIP IRES upstream of the second ORF. For example the spacer, designated “New spacer” in the tables below, has the sequence ACGTCTTCTCTTTTTAAAGGACCTCGTGAAATAAAAGTGC (SEQ ID NO:408), and the spacer, designated “Longer spacer” in the tables below, has the sequence TCCGAGCCAAGTAAGGAGGTCCCTCTCTCTCTCTCCCCCCACGTCTTCTCT TTTTAAAGGACCTCGTGAAATAAAAGTGC (SEQ ID NO:409).

These results indicate that the addition of a VCIP IRES generally increases the expression levels of the first payload encoded on a bicistronic construct, and increases the expression levels of the second payload in certain constructs. STING-Induced Type I IFN Activity, as Measured by Luminescence Levels from the IFNp-Lucia™ Reporter, and Normalized by GFP Co-Transfection, in Transfected HEK293T STING Null Cells

SD = Standard deviation

*New spacer = ACGTCTTCTCTTTTTAAAGGACCTCGTGAAATAAAAGTGC

**Longer spacer =

TCCGAGCCAAGTAAGGAGGTCCCTCTCTCTCTCTCCCCCCACGTCTTCTCTTTTTAA AGGAC CTCGTGAAATAAAAGTGC

Concentration of Expressed huIL-15Ra-IL-15sc, as Measured by ELISA and Normalized by GFP Co-Transfection, in Transfected HEK293T STING Null Cells

*New spacer = ACGTCTTCTCTTTTTAAAGGACCTCGTGAAATAAAAGTGC **Longer spacer = TCCGAGCCAAGTAAGGAGGTCCCTCTCTCTCTCTCCCCCCACGTCTTCTCTTTTTAAAGG AC CTCGTGAAATAAAAGTGC

The differences in the downstream IFN-p signaling, induced in human M2 macrophages by the different huIL-15Ra-IL-15sc and huSTINGN154S/R284G tazCTT constructs, was determined. Nucleic acids encoding huIL-15Ra-IL-15sc and huSTING N154S/R284G tazCTT were cloned into vectors, with and without a VCIP IRES, placed before the start codon of either the huIL-15Ra-IL-15sc or huSTING N154S/R284G tazCTT coding sequences.

Frozen human monocytes, isolated from healthy human donors, were thawed in complete medium (RPMI-1640 + 10% FBS), and washed by centrifugation for 10 minutes at 600xg at room temperature. The monocytes were resuspended in ImmunoCult™- SF Macrophage Medium (StemCell Technologies), containing 100 ng/mL human M-CSF. The monocytes (5e5 cells per well) then were seeded in a 24- well plate with a final volume of 500 pL. Five days later, 250 pL of ImmunoCult™- SF Macrophage Medium (StemCell Technologies), containing 300 ng/mL human M- CSF + 60 ng/mL human IL-4 + 60 ng/mL human IL-10, was added to each well, and the cells were incubated for 2 more days. On day 7, the cells were transfected using Viromer® RED mRNA and plasmid transfection reagent, according to the manufacturer’s instructions. 750 ng of plasmid DNA, from a panel of different plasmid constructs, as well as a “no DNA” control (encoding secNanoLuc®), were diluted in the provided buffer, and mixed with Viromer® RED transfection reagent, and the mixture was incubated at room temperature for 15 minutes to allow the DNA/Viromer® RED complexes to form. The DNA/Viromer® RED complexes were then slowly added to each well of the 24-well plate (in triplicate), and the plate was incubated at 37 °C in a 5% CO2 incubator. Cell culture supernatants were harvested at 48 hours, and were assayed for IFN-P using a human cytokine panel U-Plex® assay (Meso Scale Discovery), according to the manufacturer’s protocol. The average of three measurements was calculated, and background signal from the no DNA control was subtracted, to calculate the net IFN-P expression levels.

The results, which are summarized in the table below, show that the construct encoding only huIL-15Ra-IL-15sc (and no STING), does not induce any IFN-P expression in transfected human M2 macrophages. The construct that results in the highest level of IFN-P expression in transfected human M2 macrophages is the construct designated pATI2.1 CMV VCIP huIL-15Ra-IL-15sc T2A huSTING N154S/R284G tazCTT HPRE bGHpA. This construct (designated pATI2.1 CMV VCIP huIL-15Ra-IL-15sc T2A huSTING N154S/R284G tazCTT HPRE bGHpA) results in a higher level of IFN-P expression than constructs with the pATIl .76 backbone, and with no VCIP IRES at all, or with a VCIP IRES between the ORFs of huIL-15Ra-IL-15sc and huSTING N154S/R284G tazCTT (i.e., where the VCIP IRES replaces a 2 A sequence). The construct designated pATI2.1 CMV VCIP huIL-15Ra- IL-15sc T2A huSTING N154S/R284G tazCTT HPRE bGHpA results in a higher IFN-p signal, compared to the corresponding construct containing a P2A sequence instead of a T2A sequence. These results indicate that bicistronic constructs, containing a VCIP IRES upstream of the two ORFs, and a 2A sequence, particularly T2A, between the two ORFs, result in higher expression levels of the second encoded payload (in this case, a modified STING protein).

Net Signal of IFN-P Protein Expression Following Transfection of Human M2 Macrophages with Various Constructs

*New spacer = ACGTCTTCTCTTTTTAAAGGACCTCGTGAAATAAAAGTGC **Longer spacer = TCCGAGCCAAGTAAGGAGGTCCCTCTCTCTCTCTCCCCCCACGTCTTCTCTTTTTAAAGG AC CTCGTGAAATAAAAGTGC

EXAMPLE 28

Expression of Phospholipase D as a Method to Improve Plasmid Delivery

As part of its pathogenesis, Salmonella enterica serovar Typhimurium is phagocytized by myeloid cells, and activates a virulence program to alter phagosomes into Salmonella-cowt&mm^ vacuoles (SCVs). While ectopic expression of genes encoded on the plasmid requires plasmid translocation from the SCV to the nucleus, the mechanism of this translocation event is unknown. This example provides an engineering approach to improve plasmid transfer from the engineered immunostimulatory bacteria.

Rickettsia prowazekii is a Gram-negative intracellular pathogen that must escape the phagosomal compartment in order to access the host cytosol and replicate. Previous work has shown that the secreted R. prowazekii virulence factor, Phospholipase D (Pld), can be cloned into . enterica to facilitate escape from the SCV (see, e.g., Whitworth et al. (2005) Infection and Immunity 73(10):6668-6673). The pld gene (SEQ ID NO:432), encoding Pld (SEQ ID NO:433), was placed under the control of the ssaG promoter, which is induced in the SCV (see, e.g., Walthers et al. (2007) Molecular Microbiology 65(2):477-493), to ensure high expression after the bacteria is delivered into phagocytic cells. This bacterial expression cassette was cloned into the backbone of two NanoLuciferase® reporter plasmids, designated CMV-NanoLuc® and EFla-NanoLuc®, and plasmid delivery was assessed in human macrophages.

To measure plasmid delivery into infected cells, primary human monocytes were differentiated in ImmunoCult™- SF Macrophage Medium (StemCell Technologies), containing 100 ng/ml GM-CSF, for 5 days. On day 6, the cells were supplemented with additional media containing 150 ng/ml GM-CSF and 60 ng/ml huIL-4 for 48 hours. The cells were then bactofected with stationary phase strains of YS I 646Ac/.s<7 ^/ AFI.GZ pagPI i\ansF>! csgD bacteria, comprising one of the CMV- NanoLuc®, CMV-NanoLuc® + pld, EF 1 a-NanoLuc®, or EFl a-NanoLuc® + pld plasmids. The bacterial strains were grown overnight at 37 °C in 4XYT medium, at an MOI (multiplicity of infection) of 100. Infected cells were centrifuged for 5 minutes at 500 ref, and then incubated at 37 °C for 1 hour, after which the cells were washed thrice with DPBS, and incubated in fresh ImmunoCult™- SF Macrophage Medium, containing 100 pg/ml gentamicin to eliminate any extracellular bacteria. At 24 and 48 hours post-infection, the cell supernatants were harvested, and the NanoLuciferase® luciferase contents were measured using a luciferase assay (Promega). The cells also were lysed with 0.1% Triton X in PBS and plated, to enumerate the number of colony forming units (CFUs) that remained inside the cells.

The results, which are summarized in the table below, show that, when placed under the control of the CMV promoter, the levels of NanoLuciferase® expressed from the y>/<7-encoding group (i.e., the CMV-NanoLuc® luciferase + pld plasmids) were greater than the levels of NanoLuciferase® expressed from the CMV- NanoLuc® luciferase plasmid alone, at 24 hours post-infection; this difference grew in magnitude at 48 hours post-infection.

Bactofection of cells with engineered immunostimulatory bacteria encoding plda o resulted in higher luminescence levels than corresponding infections without pld, when the NanoLuc® was placed under the control of the EF-la promoter. This effect was lower than what was observed in the groups with the CMV promoter, indicating that some pld expression is enhanced by leakiness from the CMV promoter.

Comparison of Luciferase Expression Levels in Cell Supernatants of Human Macrophages Infected with Bacterial Strains Containing Plasmids Encoding Pld, vs. those not Encoding Pld ipi = hours post-infection; SD = standard deviation; RLUs = relative light units.

The results also showed that the number of CFUs of the CMV-NanoLuc® ± /^/-encoding strain, that remained inside the cells at 24 and 48 hours post-infection (hpi), decreased at a greater rate than the number of CFUs of bacteria harboring the CMV-NanoLuc® luciferase plasmid alone, indicating a different fate for each strain within the same cells.

CFUs Remaining Inside the Cells at Different Time Points

To determine if encoding Phospholipase D in the bacterial plasmid enhances the delivery of immunomodulatory payloads encoded on the same plasmid, RAW- Dual™ TLR4-KO cells (which express STING; InvivoGen) were infected with bacterial strains containing plasmids with and without nucleic acid encoding pld. RAW -Dual™ TLR4-KO cells encode a Lucia™ luciferase reporter under the control of an Interferon-Stimulated Response Element (ISRE). The cells were cultured in DMEM + 10% FBS, seeded into wells, and allowed to adhere overnight. Bactofections were performed as described above, using strains of YS I646Ac/.stZ FLG/A/ / _< g/ ^> /Ar///.sA/Ac ^, .s /J, containing plasmids encoding huIL-15Ra- ZL-15sc and huSTING N154S/R284GtazCTT, with or without nucleic acid encoding pld under control of the ssaG promoter. Uninfected cells, and uninfected cells treated with 5 pg/ml of the STING agonist 3’5’ RpRp c-di-AMP, were used as negative and positive controls, respectively. At 48 hours post-infection, the cell supernatants were harvested, and the levels of expressed Lucia™ luciferase were measured using a luciferase assay (Promega).

The results, which are shown in the table below, show that, when pld was expressed during infection, from a plasmid also encoding human IL-15Ra-IL-15sc and the STING mutant polypeptide, the ISRE activity (as measured by Lucia™ luciferase expression) was higher than in the absence of pld expression. The ISRE activity is a measure of the STING-induced type I IFN activity, and thus, is indicative of the delivery of the plasmids encoding the STING variant, and of expression of the STING variant in the cells. Expression of phospholipase D (pld), thus, enhances the delivery of plasmids that also encode immunomodulatory payloads.

Effects of pld Expression on the Delivery of Plasmids Encoding Immunomodulatory

Payload Combinations by Immunostimulatory Bacteria

STDEV = standard deviation

EXAMPLE 29

The Combination of Murine IL-15R«-IL-15sc and huSTING N154S/R284G tazCTT, and the Combination thereof with Murine Anti-CTLA-4 scFv-Fc, Induces the Secretion of CXCL10 from Myeloid Cells and Induces the Activation of T-Cells

The impact of the expression of the immunomodulatory payloads, and their combinations, on the activation and function of T-cells and myeloid cells, was evaluated. This example describes and demonstrates the impact of the delivery of various immunomodulatory payload combinations, by the immunostimulatory bacteria provided herein, on the activation of antigen-specific T-cells (in terms of CD25 expression and IFN-y secretion), and on the secretion of CXCL10, a key chemokine involved in the recruitment of anti-tumor T-cells, by myeloid cells. The levels of secreted cytokines, such as IFN-y, IFN-P, and CXCL10 (IP-10), were measured as correlates of protective anti-tumor immunity, and CD25 cell surface expression was monitored as a marker of T-cell activation. This was assessed by transfecting mouse bone-marrow derived dendritic cells (BMDCs) with plasmids encoding various combinations of payloads, co-culturing the transfected dendritic cells with autologous mouse T-cells, and then identifying the resulting cytokines.

The plasmids encoding the immunomodulatory payloads/proteins included those encoding single payloads, as well as those encoding combinations of payloads. For example, as shown in the table below, encoded payloads included murine (mu) IL-15Ra-IL-15sc (also referred to as IL-15/IL-15Ra complex, as IL-15/IL-15Ra chain complex, IL-15/IL-15R alpha chain complex, IL-15 complex, and IL-15cplex); a murine anti-CTLA-4 scFv-Fc (clone 9D9), and huSTING N154S/R284G tazCTT (a chimeric protein containing human STING with the GOF mutations N154S and R284G, and a replacement of the C-terminal tail (CTT) of human STING with the CTT of Tasmanian devil STING); and combinations thereof

Combinations of payloads included two or three payloads, expressed on a single plasmid using T2A and/or P2A peptides, such that the proteins are encoded under the control of the same promoter. The combinations of payloads included: 1) murine IL-15Ra-IL-15sc and human STING N154S/R284G tazCTT; and 2) murine anti-CTLA-4 scFv-Fc, murine IL-15Ra-IL-15sc, and human STING N154S/R284G tazCTT.

Bone marrow-derived dendritic cells (BMDCs) were differentiated from Goldenticket mice, which are STING deficient, and the cells were transfected with plasmids encoding various combinations of the investigated payloads. Twenty -four hours post-transfection, cell supernatants were harvested, and the levels of secreted CXCL10 were measured in BMDC culture supernatants, using a U-Plex® assay platform from Meso Scale Discovery, according to the manufacturer’s protocol.

To measure CD8 ⁺ T-cell activation, transfected BMDCs were pulsed with chicken ovalbumin (OVA) SIINFEKL (OVA257-264) peptide, a major histocompatibility complex (MHC) class I (H-2Kb)-restricted peptide epitope recognized by CD8 ⁺ T-cells. Splenic T-cells, isolated from RagL ^A OT-I mice, which express T-cell receptors (TCRs) that are specific for SIINFEKL presented by the MHC class I molecule H-2Kb, were added to the BMDCs for co-culture. After 24 hours of BMDC/T-cell co-culture, supernatants were harvested, and the levels of secreted IFN-y were measured using a U-Plex® assay platform from Meso Scale Discovery, according to the manufacturer’s protocol.

The co-cultured cells also were harvested, and CD8 ⁺ T-cells were stained with a phycoerythrin (PE)-conjugated murine anti-CD25 antibody (clone PC61, BioLegend), to determine the expression levels of the CD25 T-cell activation marker.

The results are summarized in the table below, which shows the levels of CXCL10 secreted by BMDCs in response to transfection with plasmids encoding various single and combination payloads, as well as the levels of IFN-y secreted by CD8 T-cells, and the levels of T-cell activation (in terms of CD25 expression), following co-culture with the transfected BMDCs.

The results show that huSTING N154S/R284G tazCTT alone induces very high levels of CXCL10 secretion by BMDCs. Combinations of murine IL-15Ra-IL- 15sc with huSTINGN154S/R284G tazCTT, or of murine anti-CTLA-4 scFv-Fc + murine IL-15Ra-IL-15sc + huSTING N154S/R284GtazCTT, also induce high levels of CXCL10 secretion by BMDCs.

The results also show exemplary combinations of encoded payloads that induce high levels of secretion of IFN-y by human T-cells, and induce the activation (CD25 expression) of T-cells. An increased effect was observed with the combination of muIL-15Ra-IL-15sc and huSTINGN154S/R284G tazCTT. The combination of muIL-15Rci-IL-15sc and huSTINGN154S/R284G tazCTT also results in increased activation of CD8 ⁺ T-cell responses, and the expression of CD25. These results indicate that the delivery of the combination of IL-15Ra-IL-15sc and the modified STING chimera with a gain-of-function mutation or mutations and with a CTT replacement to reduce NF-KB signaling, to the tumor microenvironment, increases advantageous anti-tumor immune responses, and reduces undesirable inflammatory responses.

SEM = Standard Error of the Mean

Delivery of this combination of encoded payloads, by the immunostimulatory bacteria provided herein, or by other delivery vehicles, such as oncolytic viruses or vectors, for expression of the payloads in tumor-resident myeloid cells and/or in the tumor microenvironment, provides for increased anti-tumor responses in the treated subject. As also shown in Examples below, the combination of a cytokine and a modified STING protein with a gain-of-function mutation, including the chimeras, results in additional advantageous anti -tumor responses in a treated subject.

EXAMPLE 30

Human Recombinant IL-15, in Combination with a Small Molecule STING Agonist, Induces the Secretion of CXCL10 from Myeloid Cells, and Induces the Activation of T-Cells

This Example shows the impact of the expression of human immunomodulatory payloads, and their combinations, on the activation and function of T-cells and myeloid cells. The levels of secreted cytokines, such as IFN-y and CXCL10 (IP- 10), were measured as indicators of protective anti -tumor immunity. This Example describes and demonstrates the impact of the recombinant monomeric human IL-15 cytokine and 2’3’-c-di-AM(PS)2 (Rp, Rp), a small molecule STING (smSTING) agonist, on the secretion of CXCL10 by myeloid cells, a chemokine that is involved in the recruitment of anti-tumor T-cells, and on the secretion of IFN-y by antigen-specific T-cells. The smSTING agonist is a cyclic dinucleotide (CDN) known to induce the production of type I interferons (IFNs) following its recognition by endoplasmic reticulum-resident STING. 2'3'-c-di-AM(PS)2 (Rp, Rp) is the Rp, Rp- isomer of the 2'3'-bisphosphorothioate analog of 3 '3 '-cyclic adenosine monophosphate (c-di-AMP). It has a higher affinity for STING than c-di-AMP, due to the presence of a 2'-5', 3'-5' mixed linkage. This analog contains two phosphorothioate diester linkages to protect it against degradation by phosphodiesterases that are present in host cells or in the systemic circulation. 2'3'-c-di-AM(PS)2 (Rp, Rp) strongly induces type I IFN and CXCL10 production in relevant cells.

Human monocyte-derived dendritic cells (ModDCs) and autologous human T- cells were co-cultured with the small molecule (smSTING) agonist and the recombinant IL-15, and the resulting secreted cytokines were identified. The human monocyte-derived dendritic cells (ModDCs) were differentiated from negatively isolated monocytes, using the ImmunoCult™ Dendritic Cell Culture Kit (STEMCELL Technologies), for 6 days, following the manufacturer’s instructions. After 6 days of differentiation, ModDCs were pulsed with a pool of 32 MHC class I- restricted viral peptides from human cytomegalovirus (CMV), Epstein-Barr virus (EBV), and influenza virus (Flu) (extended CEF peptide pool for human CD8 T cells, MABTECH), before the addition of autologous CD8 T-cells, and treatment with 2.5 nM of recombinant human IL-15 and/or 5 pg/ml of smSTING. Pulsing the dendritic cells with the viral peptides causes the dendritic cells to present the viral antigens on their cell surfaces, and stimulates antigen-specific T-cells to produce IFN-gamma. After 48 hours of ModDC/T-cell co-culture, the levels of secreted IFN-y and CXCL10 were measured in the cell culture supernatants, using a U-Plex® assay platform from Meso Scale Discovery, according to the manufacturer’s protocol.

The results, which are summarized in the table below, show the additive effect of STING activation and human IL- 15 activity on the secretion of CXCL10 by human dendritic cells upon antigen-specific stimulation. The results also show that the combination of STING activation and a cytokine, such as human IL-15, synergistically activates antigen-specific CD8 ⁺ T-cells, and induces high levels of IFN-y secretion by the activated CD8 ⁺ T-cells.

SEM = Standard Error of the Mean

EXAMPLE 31 Combinations of Gain-of-Function STING Variants and Cytokines Enhance the Anti-Tumor Immune Response

The Combination of Human IL-15Ra-IL-15sc and huSTING N154S/R284G tazCTT Induces the Secretion of IFN-P from Myeloid Cells, and Induces the Activation of T-Cells

This Example demonstrates the impact of the encoded human immunomodulatory payloads, and their combinations, that are delivered for expression in the TME and/or in tumor-resident myeloid cells, on the activation and function of human T-cells and dendritic cells. The levels of cytokines secreted by the dendritic cells and the T-cells, such as IFN-y and IFN-P, respectively, were measured as correlates of protective anti-tumor immunity. This example describes and demonstrates the impact of the delivery of various immunomodulatory payload combinations to the tumor microenvironment, such as by the immunostimulatory bacteria exemplified herein, and/or by other delivery vehicles described herein, on the activation of antigen-specific T-cells (as demonstrated by IFN-y secretion), and on the secretion of IFN-P, a key factor involved in the anti-tumor immune response. This was achieved by transfecting human monocyte-derived dendritic cells (ModDCs) with plasmids encoding various combinations of payloads, co-culturing the transfected dendritic cells with autologous human T-cells, and identifying the cytokines secreted as a result. The plasmids encoding the immunomodulatory payloads/proteins included those encoding single payloads, and those encoding combinations of payloads. For example, as shown in the table below, the encoded payloads included huIL-15Ra-IL- 15sc alone, huSTING N154S/R284GtazCTT alone, and the combination ofhuIL- 15Ra-IL-15sc and huSTINGN154S/R284G tazCTT.

Human monocyte-derived dendritic cells (ModDCs) were differentiated from negatively isolated monocytes using the ImmunoCult™ Dendritic Cell Culture Kit (STEMCELL Technologies) for 6 days, following the manufacturer’s instructions. After 6 days of differentiation, ModDCs were transfected with plasmids encoding the various investigated payloads and combinations thereof, using the Viromer® RED mRNA and plasmid DNA transfection reagent (Origene). Four hours posttransfection, the ModDCs were pulsed with a pool of 32 MHC class Lrestricted viral peptides from human cytomegalovirus (CMV), Epstein-Barr virus (EBV), and influenza virus (Flu) (extended CEF peptide pool for human CD8 T cells, MABTECH), before the addition of autologous CD8 ⁺ T-cells to the cell culture. The transfected dendritic cells also were pulsed with an irrelevant peptide from the Human Immunodeficiency virus 1 (HIV-1) reverse transcriptase, and used as a negative control for the peptide stimulation of T-cells. CD8 ⁺ T-cells from HIV-1 negative donors were used. After 48 hours of ModDC/T-cell co-culture, the levels of IFN-y secreted by the T-cells were measured in the cell culture supernatants, using a TIPI ex® assay platform from Meso Scale Discovery, according to the manufacturer’s protocol.

In a separate experiment, ModDCs were transfected with plasmids encoding the various investigated payloads and combinations thereof, using the Viromer® RED mRNA and plasmid DNA transfection reagent (Origene), and cultured for forty-eight hours. Cell culture supernatants from the transfected ModDCs were harvested fortyeight hours post-transfection, and the levels of secreted IFN-P were measured using a U-Plex® assay platform from Meso Scale Discovery, according to the manufacturer’s protocol. In order to analyze the impact of the encoded payloads on the secretion of IFN-P by the human dendritic cells, the concentration of IFN- measured for the plasmid control group (Beta-actin) was subtracted from all the other groups in the study, providing the net IFN-P concentrations resulting only from the activities of the encoded payloads.

The results, which are summarized in the table below, show the synergistic effect of the combination of huIL-15Ra-IL-15sc and huSTING N154S/R284G tazCTT activities on the secretion of IFN-P by human dendritic cells. The results also show the combined effects of huIL-15Ra-IL-15sc and huSTING N154S/R284G tazCTT, which are at least additive, on the activation of antigen-specific CD8 ⁺ T- cells, and the induction of high levels of secreted IFN-y from the activated T-cells.

SEM = Standard Error of the Mean

It is shown in the next Example, that the combined effects of the encoded payloads lead to synergistic results, as evidenced by the cures observed in mouse models of cancer.

EXAMPLE 32

Combinations of muIL-15Ra-IL15sc, huSTING N154S/R284G tazCTT, and Murine Anti-CTLA-4 scFv-Fc, Demonstrate Superior Anti-Tumor Efficacy in a

Highly Refractory Mouse Model of Triple Negative Breast Cancer

The synergistic effects observed in the previous Example (Example 31), are further shown in this Example, which shows that the combinations of payloads act synergistically in effecting cures. Various payload combinations were evaluated for in vivo efficacy in an orthotopic, T-cell excluded, and metastatic model of CPI (immune checkpoint inhibitor) refractory triple-negative breast cancer. For this experiment, 6-8 week-old female BALB/c mice (8 mice per group) were inoculated in the left mammary fat pad with EMT6 tumor cells (ATCC® CRL-2755™) (IxlO ⁶ cells in 100 pL PBS). Mice bearing 7 day-old established mammary tumors (~65 mm ³ in volume) were intravenously (IV) injected with a single dose of 1 xlO ⁷ CFUs of strains of YS1646Aasz//AFLG/A/?tzgP/Aaws _J 8/AcsgD containing plasmids encoding the various payloads, alone, or in combination with weekly intraperitoneal (IP) injections of 100 pg of the anti-PD-Ll antibody atezolizumab. The YS1646Aasz//AFLG/A/?tzgP/Aaws _J S/AcsgD strains contained plasmids encoding muIL- 15Ra-ILl 5 sc + huSTING N154S/R284G tazCTT, or muIL-15Ra-ILl 5 sc + huSTING N154S/R284G tazCTT + muAnti-CTLA-4 scFv-Fc, or the control plasmid encoding P-actin, and were compared to treatment with PBS control.

The results are summarized in the table below. The tumors in the PBS-treated mice grew evenly, reaching a max tumor volume at day 31. Treatment of mice with the control p-actin plasmid did not demonstrate much evidence of anti-tumor efficacy (5% tumor growth inhibition (TGI), 2/8 cures), nor did treatment with IP anti-PD-Ll alone (17.9% TGI, 1/8 cures). Mice that were IV treated with the bacterial strain containing a plasmid encoding muIL-15Ra-IL15sc + huSTING N154S/R284G tazCTT demonstrated monotherapy efficacy (59.8% TGI, 3/8 cures), which improved with the combination of IP anti-PD-Ll (77.3% TGI, 5/8 cures). The combination of muIL-15Ra-IL15sc + huSTING N154S/R284G tazCTT + muAnti-CTLA-4 scFv-Fc also demonstrated significant monotherapy efficacy (62.4% TGI, 4/8 cures).

The therapeutic payload combinations were very well tolerated, and mice did not lose weight during the study. These data demonstrate the in vivo potency of IV administered immunostimulatory bacteria encoding the combinations of muIL-15Ra- IL15sc + huSTING N154S/R284G tazCTT, alone, or enhanced with IP administered anti-PD-Ll, and muIL-15Ra-IL15sc + huSTING N154S/R284G tazCTT + muAnti- CTLA-4 scFv-Fc, in an orthotopic, T-cell excluded, and metastatic model of checkpoint inhibitor refractory triple-negative breast cancer (TNBC).

Effects of Combinations of Encoded Payloads, Delivered by Immunostimulatory Bacteria, with and without addition of IP Anti-PD-Ll, on the Inhibition of Tumor Growth and on the Cure Rates of Mice with Checkpoint Inhibitor Refractory Triple-Negative Breast Cancer

A follow-up study was performed to evaluate the efficacy of higher doses of the muIL-15Ra-IL15sc + huSTING N154S/R284G tazCTT combination therapy, and for comparison with the efficacies of individually administered muIL-15Ra-IL15sc and huSTING N154S/R284G tazCTT. Strains of N PA6 asdl AFLG/ pagPI ansBI csgD containing plasmids encoding muIL- 15Ra-IL15sc, huSTINGN154S/R284G tazCTT, or the combination thereof, were administered intravenously (IV), at a dose of 3e7 CFUs, to mice bearing 7 day-old established mammary tumors, and were compared to PBS control. The tumors in the PBS-treated mice grew evenly, reaching a maximum tumor volume at day 31. As shown in the table below, mice treated with the bacterial strain containing a plasmid encoding muIL-15Ra-IL15sc or huSTING N154S/R284G tazCTT alone demonstrated 2/10 cures each, whereas the combination of muIL-15Ra-IL15sc + huSTING N154S/R284G tazCTT resulted in 7/10 cures.

These data show the synergistic potency of combinations of a cytokine and a modified STING protein, such as the combination of muIL-15Ra-IL15sc + huSTING N154S/R284G tazCTT, for promoting complete responses. It is particularly significant, since these results were achieved in a highly refractory model of breast cancer, which highlights the general anti-tumor therapeutic potency of the combination. These results indicate the synergistic anti-tumor potency of the combination of a cytokine, such as the IL-15Ra-IL15sc fusion protein, and a STING protein, particularly a highly active STING protein, such as a gain-of-function constitutively active STING variant, or a gain-of-function constitutively active STING variant that is further modified to have reduced NF-KB signaling.

Cured mice from the muIL-15Ra-IL15sc + huSTING N154S/R284G tazCTT combination treatment groups (described above) then were evaluated for the ability to promote durable anti -tumor immunity in an orthotopic EMT6 tumor re-challenge study, and in a CD8 ⁺ T-cell dependent manner. For this study, 20 cured mice were divided into two groups of 10 mice each, and each mouse received IP injections of either 100 pg of an anti-CD8p antibody (which does not deplete CD8a ⁺ dendritic cells), or 100 pg of an IgG isotype control, on days 3 and 1 prior to tumor rechallenge (day 56 and day 58 post-initial tumor implantation, respectively). The mice were bled prior to tumor re-challenge to confirm CD8 ⁺ T-cell depletion, and average circulating CD8 ⁺ T-cells were determined to be 5.72% for the isotype control, and 0.48% for the anti-CD8p antibody. The mice were then re-challenged with le6 EMT6 tumor cells, orthotopically on the opposite mammary fat pad, and compared to naive, age-matched control mice (N=10). Mice from the naive group all grew tumors to maximum tumor volume by day 30. Re-challenged mice from the anti-CD8p antibody-depleted group grew tumors even more aggressively than the naive mice, whereas all 10 re-challenged mice from the IgG isotype antibody control group (i.e., mice without CD8 ⁺ T-cell depletion) were protected from tumor re-challenge. These data demonstrate the significant and durable anti-tumor efficacy of the combination of a cytokine and a modified gain-of-function constitutively active STING protein variant, such as when delivered in an immunostimulatory bacterium, such as a strain of YS1646Acz5t7/AFLG/A >cz /Atzn5B/Ac5gD bacteria, comprising plasmids encoding a combination of muIL-15Ra-IL15sc + huSTING N154S/R284GtazCTT. The combination of therapeutic payloads/proteins induces a high cure rate after IV administration of the bacteria, and mice are protected from tumor re-challenge in a CD8 ⁺ T-cell dependent manner.

EXAMPLE 33

DLL3xCD3 T-Cell Redirection Antibody (TCRA) Design

T-cell redirection antibodies (TCRAs), or T-cell redirecting bispecific antibodies (TRBAs), are T-cell receptor (TCR) function-independent T-cell based immunotherapies, where the epsilon (e) domain of cluster of differentiation 3 (CD3), a component of the TCR complex, is targeted by one binding domain, while a second binding domain targets a tumor cell surface antigen. TCRAs include, for example, bispecific T-cell engagers (sold under the trademark BiTEs®), which are engineered antibody-based immunotherapies, that are fusion proteins of two single-chain variable fragments (scFvs), linked together by a peptide linker.

TCRAs induce the formation of a cytolytic synapse between the T-cell and the tumor cell, in an MHC-independent and TCR function-independent manner, by simultaneously binding to a tumor cell surface antigen, and to CD3 on T-cells, directing the cytolytic activity of T-cells selectively to the targeted tumor cells. Following the formation of the cytolytic synapse, the T-cells release cytotoxic proteins, such as perforin and granzymes (e.g., granzyme B), resulting in the apoptosis of the tumor cells. The activation of the T-cells results in the release of cytokines, engaging other immune cells and inducing a broader anti-tumor immune response. This results in the conversion of a non-inflamed (or cold) tumor environment to an inflamed (or hot) one, and results in the infiltration and proliferation of T-cells, and the killing of tumor cells. The original BiTE® antibodies have a short half-life; later generation BiTE® antibodies have an extended half-life by virtue of fusion to an Fc (see, e.g., Hipp et al. (2020) Clin. Cancer Res. 26:5258-5268; and Strohl, W. R. and Naso, M. (2019) Antibodies 8:41).

Delta-like ligand 3 (DLL3) is an inhibitory Notch pathway ligand that plays a role in Notch signaling during embryonal development. DLL3 is expressed intracellularly during embryonal development in normal tissues, with the highest expression in the fetal brain, but is absent in adult normal tissues. DLL3 is overexpressed in certain tumors, such as small cell lung cancers (SCLCs) and other high-grade neuroendocrine tumors, where, instead of being expressed intracellularly, it escapes to the cell surface, making it amenable to targeting with antibody-based therapeutics (see, e.g., Hipp etal. (2020) Clin. Cancer Res. 26:5258-5268). DLL3 also is expressed in other tumor types of neuroendocrine origin, including melanoma, glioblastoma multiforme, small cell bladder cancer, metastatic castration-resistant prostate cancer, and neuroendocrine lung tumors (see, e.g., Owen etal. (2019) Journal of Hematology & Oncology 12:61); hence, it is a useful target for TCRA- based therapies.

A DLL3xCD3 TCRA (see, SEQ ID NOs: 410 and 411 for the nucleotide and amino acid sequences, respectively; see, also, Hipp et al. (2020) Clin. Cancer Res. 26:5258-5268) was prepared from the light chain and heavy chain variable regions (VL and VH, respectively) of the anti-DLL3 clone hSC16.56 antibody (see, e.g., International Application Publication No. WO 2013/126746), and the light chain and heavy chain variable regions of the anti-CD3e clone 145-2C11 antibody (available from BioLegend).

The construct included the mouse IgGK leader/signal sequence from the pSecTag2 vector (corresponding to residues 1-21 of SEQ ID NO:411; see, e.g., addgene.org) for secretion of the TCRA. The anti-DLL3 VH domain (corresponding to residues 22-139 of SEQ ID NO:411) was linked to the anti-DLL3 VL domain (corresponding to residues 155-261 of SEQ ID NO:411) via a 15 amino acid long (Gly4Ser)3 linker (corresponding to residues 140-154 of SEQ ID NO:411), forming an anti-DLL3 scFv. A 5 amino acid long Gly4Ser linker (corresponding to residues 262- 266 of SEQ ID NO:411) was inserted between the anti-DLL3 scFv and the anti-CD3 scFv. The anti-CD3 scFv contains the anti-CD3 VH domain (corresponding to residues 267-382 of SEQ ID NO:411), linked, via a second (Gly4Ser)3 linker (corresponding to residues 383-397 of SEQ ID NO:411), to the anti-CD3 VL domain (corresponding to residues 398-504 of SEQ ID NO:411). This construct was designated DLL3HL x CD3HL, to indicate the order in which the heavy (H) and light (L) chain variable domains of the anti-DLL3 and anti-CD3 antibodies appear in the molecule. A FLAG tag (DYKDDDDK; SEQ ID NO:412) can be added to the C-terminal end of this construct for detection (for the FLAG tag-labeled DLL3xCD3 TCRA, designated DLL3HL x CD3HL-FLAG TCRA, see SEQ ID NO:413)

Another DLL3xCD3 TCRA construct, designated DLL3LH x CD3HL TCRA (see, SEQ ID NOs: 414 and 415 for the nucleotide and amino acid sequences, respectively), was constructed similarly to the DLL3HL x CD3HL construct, but, as the designation implies, the anti-DLL3 scFv contains the anti-DLL3 VL domain placed N-terminally to the anti-DLL3 VH domain. The anti-CD3 scFv portion of the TCRA, the GS linkers, and the leader sequence for secretion, all are as described above for the DLL3HL x CD3HL construct. A FLAG tag-labeled version of the DLL3LH x CD3HL TCRA construct, designated DLL3LH x CD3HL-FLAG TCRA (see, SEQ ID NOS: 416 and 417 for the nucleotide and amino acid sequences, respectively), also was constructed. EXAMPLE 34 Evaluation of T-Cell Redirection Antibodies (TCRAs) For Target Binding, Conjugation, And Cytotoxicity

Antibodies directed against CD (cluster of differentiation) antigens, such as CD19 and CD3, are anti-cancer targets. Expression of certain CD antigens is restricted to specific lineage lymphohematopoietic cells. Antibodies directed against lymphoid-specific antigens have been developed as therapeutics. CD 19 is a useful target because it is expressed on B cells, it is not shed, it is uniformly expressed on all lymphoma cells, and it is absent from stem cells. CD3 is expressed on T-cells as part of the T-cell receptor (TCR) complex, which contains three chains: CD3e, CD36, and CD3y. The clustering of CD3 on T-cells, such as by immobilized anti-CD3 antibodies, results in T-cell activation that is similar to engagement of the T-cell receptor (TCR), but is independent of its clone typical specificity. A bispecific antibody, directed against each of the CD19 and the CD3 antigens, retargets T-cell cytotoxicity towards CD-19 positive cells (e.g., lymphoma cells), in an MHC- independent manner, and without the need for T-cell pre- stimulation or costimulation. Thus, such a molecule is particularly useful in inducing durable T-cell mediated anti-tumor immunity, and as an anti-cancer therapeutic.

The FLAG-tag variant sequence of the anti-CD19 and anti-CD3 TCRA, designated MT103 (SEQ ID NO:418; see, also, expired U.S. PatentNo. 7,112,324), was cloned into a vector designated pATI1.75 (described above). The MT103 TCRA construct contains a leader sequence (corresponding to residues 1-19 of SEQ ID NO:418), followed by a FLAG tag (DYKDDDDK; corresponding to residues 20-27 of SEQ ID NO:418), and an anti-CD19 scFv linked to an anti-CD3 scFv. The anti- CD19 scFv contains an anti-CD19 VL domain (corresponding to residues 28-138 of SEQ ID NO:418), linked to an anti-CD19 VH domain (corresponding to residues 154- 277 of SEQ ID NO:418), via a (GGGGS)3 linker (corresponding to residues 139-153 of SEQ ID NO:418). The anti-CD19 scFv is linked to the anti-CD3 scFv via a GGGGS linker (corresponding to residues 278-282 of SEQ ID NO:418). The anti- CD3 scFV contains an anti-CD3 VH domain (corresponding to residues 283-401 of SEQ ID NO:418), linked via a peptide linker (VEGGSGGSGGSGGSGGVD; corresponding to residues 402-419 of SEQ ID NO:418), to an anti-CD3 VL domain (corresponding to residues 420-525 of SEQ ID NO:418). The construct also includes a His tag (Hise; corresponding to residues 526-531 of SEQ ID NO:418).

HEK293T STING Null Cells (293 -Dual™ Null Cells; InvivoGen) were seeded in 6-well plates coated with poly-L-lysine at 1.5e6 cells per well, and incubated overnight at 37 °C in a 5% CO2 incubator, to achieve 80% confluency. The following day, 3000 ng of the MT103 plasmid DNA was diluted in serum-free media, and added to FuGENE® transfection reagent (Promega), at the proper reagent:DNA ratios, and the cells were transfected, with untransfected wells serving as negative controls (in duplicates). Cell culture supernatants from each sample were collected 48 hours post-transfection. The supernatant was either kept neat, or was concentrated using an Amicon® Ultra 4 mL centrifugal filter (Millipore Sigma).

To confirm the functionality of the expressed MT103 (CD19xCD3) TCRA, a flow cytometry experiment using Raji cells (purchased from ATCC) and Jurkat- Lucia™ NF AT cells (InvivoGen) was performed. The Raji cells are human lymphoblast cells that highly express CD 19, and Jurkat cells are human T- lymphocytes that express CD3. 200,000 Raji cells, and 200,000 Jurkat cells, each were washed in PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes. Then, 50 uL of either neat or concentrated cell culture supernatant, containing the expressed TCRA, was added to the cells, and incubated for 30 minutes at 37 °C in a 5% CO2 incubator. The cells then were resuspended in PBS + 2% FBS, and stained with an APC-labeled anti-FLAG tag antibody (BioLegend). The APC-labeled anti-FLAG tag antibody is added to label any MT103 TCRA (which contains a FLAG tag) that binds to CD19 on the Raji cells, and/or CD3 on the Jurkat cells. After a 30 minute incubation, the cells were washed twice with PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes, and then were resuspended in PBS + 2% FBS with DAPI (dead/live stain). Flow cytometry was performed, to detect any binding of the CD19xCD3 MT103 TCRA to CD 19 on the Raji cells, and/or binding of the TCRA to CD3 on the Jurkat cells. 50,000 events were acquired for each sample. Flow cytometry data were acquired using the ACEA NovoCyte® flow cytometer (ACEA Biosciences, Inc.) and analyzed using the FlowJo™ software (Tree Star, Inc.).

As shown in the table below, the median fluorescence intensity (MFI) of APC detected from the Raji and the Jurkat cells that were incubated with either neat or concentrated MT 103 -containing cell culture supernatant, is much higher than the MFI detected from the cells incubated with untransfected cell culture supernatant, indicating that the MT103 TCRA binds to the CD19- and CD3 -expressing cells.

Flow Cytometry Data for Cells Treated with the MT103 (CD19xCD3) TCRA

Additionally, the CD19xCD3 MT 103 TCRA construct was tested in a coculture of Raji and Jurkat-Lucia™ NF AT cells. The Jurkat-Lucia™ NF AT reporter cell line contains an NF AT -inducible Lucia™ construct, where expression of Lucia™ luciferase is driven by an ISG54 minimal promoter, fused to six copies of the NF AT (nuclear factor of activated T-cells; a transcription factor) consensus transcriptional response element. The Jurkat-Lucia™ NF AT cells, thus, can be used to study NF AT activation by monitoring Lucia™ luciferase activity; the levels of Lucia™ luciferase can be measured in the cell culture supernatant with QUANTI-Luc™, a Lucia™ luciferase detection reagent. In this experiment, the binding of the MT103 TCRA to both the CD19-expressing Raji cells, and the CD3 -expressing Jurkat cells, activates the Jurkat cells (T-cells), resulting in NF AT -inducible Lucia™ luciferase secretion, which can be measured to determine the level of activation of the Jurkat cells.

Jurkat cells were seeded at 5e4 cells per well in a 96-well tissue culture plate. Raji cells were seeded at either le5 or 5e4 cells per well, for an Effector: Target (Jurkat:Raji) ratio of 1:2 or 1 :1, respectively. The co-cultured cells were incubated with and without the neat or concentrated MT 103 -containing cell culture supernatant. 10-fold, 100-fold, and 1000-fold dilutions of the neat or concentrated supernatant were prepared and incubated with the co-cultures. The luminescence (z'.e., secreted Lucia™ luciferase activity) resulting from the induced NF AT reporter was detected at 6 hours and 24 hours post-addition of the TCRA to the co-cultures.

The results are summarized in the table below, which shows that the activity of the MT 103 TCRA results in increased luminescence from the Jurkat-Lucia™ NF AT reporter cell line, compared to the control samples without the addition of MT103, and to the control samples without the co-culture with Raji (target) cells. The luminescence resulting from the NF AT -inducible Lucia™ reporter construct decreases with the dilution of MT103, in a dose-dependent manner. A longer incubation time (24 hours vs. 6 hours) results in increased luminescence, as expected, which is indicative of increased T-cell activation.

NFAT-Lucia™ Reporter Values from Co-Cultures of Jurkat and Raji Cells Treated with Different Concentrations of MT103, and with Different Effector: Target Ratios

SD = standard deviation

A cytotoxicity assay was performed using a co-culture of human CD8 ⁺ T-cells and Raji cells (Burkitt’s lymphoma cells; i.e., tumor cells). Human CD8 ⁺ T-cells were isolated using the EasySep™ Human T-cell Isolation Kit (StemCell Technologies), and then were resuspended in RPMI + 2% FBS, and seeded in a 96-well plate. CD8 T-cells were seeded at le5 cells per well, or at 5e4 cells per well, and Raji cells were seeded at 10,000 cells per well, for Effector: Target (T-cell:Raji cell) ratios of 10: 1 or 5: 1, respectively. Concentrated or neat MT 103 -containing supernatant was diluted 10- fold, 100-fold, and 1000-fold, in RPMI + 2% FBS, and then 50 pL of each supernatant sample was added to the co-cultures. The supernatant was collected after 5 hours of co-culture incubation, and analyzed using the CytoTox 96® NonRadioactive Cytotoxicity Assay (Promega), per the manufacturer’s instructions. The % cytotoxicity was calculated according to the manufacturer’s instructions.

As shown in the table below, higher amounts (concentrations) of MT103 result in higher % cytotoxicity. The % cytotoxicity decreases as the MT 103 -containing supernatant is diluted 10-fold, 100-fold, and 1000-fold, in a dose dependent manner.

Lysis of Raji Cells (% Cytotoxicity) in Co-Culture with CD8 ⁺ T-Cells and Varying Concentrations of MT103

SD = Standard deviation; NA = not applicable

EXAMPLE 35

Immunostimulatory Bacterial Delivery of Tumor-Associated Antigens Encoded in Expression Plasmids

This Example describes exemplary tumor-associated antigens (TAAs) that can be delivered by the immunostimulatory bacteria. Upon treatment with the engineered immunostimulatory bacteria, the delivery of tumor-associated antigens to tumorresident myeloid cells, such as macrophages and dendritic cells, allows for the priming and activation of antigen-specific T-cells to result in an anti-tumor response.

Among the tumor-associated antigens encoded by the plasmids in the immunostimulatory bacteria are those categorized as oncofetal or oncoviral. Some of the antigens are either overexpressed by, or they accumulate in tumors. The antigens include those that are lineage-restricted, mutated, or post-translationally altered. Nonlimiting examples of tumor-associated antigens in each of these categories, and their associated cancer types (indications), are set forth in the following table.

EXAMPLE 36

Ectopic Protein Synthesis in Eukaryotic Cells from Bacterial-Derived RNA Delivery As described in the detailed description, in addition to plasmid DNA delivery, immunostimulatory bacterial can be used to directly deliver bacterial-derived RNA, such as mRNA, into infected myeloid cells. Direct delivery of RNA obviates the need for plasmid DNA delivery, which relies on the eukaryotic transcription and translation machinery, as well as proper trafficking of plasmid DNA to the host cell nucleus. Direct delivery of RNA, such as mRNA, bypasses host trafficking and transcriptional processes, to rely only on eukaryotic translation machinery for protein production. The delivery of bacterial-derived RNA requires the de-coupling of bacterial transcription from bacterial translation, in order to prevent problematic translation of foreign proteins in the bacteria. Many eukaryotic proteins are complex proteins that require proper folding conditions, an endoplasmic reticulum, proper post-translational modifications, including disulfide bonds, and host chaperones for folding. Overexpression of foreign proteins in bacteria can reduce strain fitness, decrease doubling time, and reduce strain stability.

To demonstrate that mRNA delivery occurs during treatment with the immunostimulatory bacteria, a series of plasmids were engineered with either eukaryotic or bacterial promoters driving the expression of mu4-lBBL_T2A_muIL- 12p70 HPRE bGHpA, with the presence and absence of an intervening encephalo- myocarditis virus Internal Ribosome Entry Site (EMCV IRES). The EMCV IRES, which is non-functional in bacteria, reduces prokaryotic translation and enhances eukaryotic translation. There is only one known eukaryotic IRES, from the Dicistroviridae family of viruses, that functions in bacteria (see, e.g., Colussi el al. (2015) Nature 519(7541): 110-113). Thus, in general, IRES elements do not function in bacteria. The promoter-IRES (or promoter-Kozak) combinations also were cloned into an mCherry fluorescent reporter gene (Takara Bio), so that bacterial expression from the regulatory sequences could be rapidly assessed by fluorescence.

To measure the bacterial expression of the fluorescent mCherry protein, overnight cultures of bacterial strains were washed in PBS, and normalized by ODeoo to 0.1. For every bacterial culture, the percentage of mCherry expression was analyzed by flow cytometry using the NovoCyte® Flow Cytometer (ACEA Biosciences, Inc.). The level of ectopic gene expression after bactofection, using a coculture system of infected primary human M2 macrophages and HEK-Blue™ IL- 12 reporter cells (InvivoGen), was measured. HEK-Blue™ IL-12 cells are HEK293 cells that express the IL- 12 receptor, as well as genes in the IL- 12 signaling pathway, and a STAT4-inducible SEAP reporter gene. The binding of IL-12 to the IL-12 receptor on the surface of HEK-Blue™ IL- 12 cells triggers a signaling cascade leading to the activation of STAT-4, with the subsequent production of SEAP. Detection of SEAP in the supernatant of HEK-Blue™ IL-12 cells can be readily assessed using QUANTI- Blue™ Solution. Thus, these reporter cells are used to measure the expression of IL-12.

Primary human monocytes were differentiated in ImmunoCult™- SF Macrophage Medium, containing 100 ng/ml M-CSF, for 3 days, and then supplemented with additional medium containing 200 ng/ml M-CSF, 20 ng/ml IL-4, and 20 ng/ml IL- 10, for 4 more days. The cells were then infected with stationary phase immunostimulatory bacteria, which were grown overnight at 37 °C in 4XYT medium, at an MOI of 50. The cells were inoculated with bacteria and centrifuged for 5 minutes at 500 ref, followed by an incubation at 37 °C for 1 hour. The cells were then washed twice with DPBS, and then incubated in fresh ImmunoCult™-SF Macrophage Medium, containing 100 pg/ml gentamicin, to remove extracellular bacteria. Infected macrophages were then co-cultured with the HEK-Blue™ IL- 12 reporter cells, at a ratio of 2.5:1 (macrophages to HEK-Blue™ cells). At 48 hours post-infection, 5 pl of cell culture supernatants were incubated with 180 pl of QUANTLBlue™ solution (which detects SEAP), and incubated at 37 °C, to allow for the color to develop. Thus, the expression of IL-12 by the infected macrophages is assessed by measuring the SEAP levels in the supernatants, which is produced by the HEK-Blue™ cells upon binding of IL-12 to the IL-12 receptor on their cell surfaces.

The results, which are summarized in the Table below, show that the infected control (EF-la-NanoLuc® luciferase) stimulates a baseline level of IL-12 expression in primary human M2 macrophages, and very little fluorescence from mCherry expression in the bacteria. The CMV-Kozak pair drives a higher level of IL-12 expression after bactofection, as well as a higher level of mCherry protein expression. This is attributable to the low levels of bacterial “leakiness” that occur from the CMV promoter in bacteria. The two bacterial promoters that were tested, rpsM and MTL, when paired with a Kozak sequence, induce levels of IL- 12 expression in primary human M2 macrophages that are lower than the CMV promoter, but higher than EF- la-NanoLuciferase® control, and both promoters drive a high level of mCherry protein expression in bacteria. These results indicate that mRNA delivery is occurring, because the bacterial promoters are non-functional in eukaryotic hosts. The levels of expression of the IL-12 reporter protein are higher when the bacterial promoters are coupled with an IRES, indicating that protein expression from delivered mRNA is enhanced at the translational level by the presence of a eukaryotic IRES. The presence of an IRES also reduces the amount of mCherry expression in the bacteria, indicating that the secondary structure of the RNA inhibits expression even when coupled with a bacterial promoter.

Including an IRES in the constructs, downstream from the promoter and before a start codon, inhibits or prevents translation in bacteria, so that the bacteria serve as delivery agents for mRNA. The bacteria, such as the immunostimulatory bacteria provided in this application, are designed to infect host cells, such as tissueresident macrophages (e.g., tumor-resident myeloid cells), and to deliver the encoded RNA into the host cells. The bacteria can auxotrophic, such as asd, so that they can be grown in vitro, but such that they do no grow when introduced into a host, such as a human. The bacteria deliver their contents to host cells, and do not reproduce.

Bacteria are cultured in vitro to produce encoded RNA, which, because of the presence of an IRES or other such regulatory sequence that prevents or inhibits bacterial translation, results in the production of RNA, such as mRNA by the bacteria. Upon infection of eukaryotic host cells, such as human macrophages, by the bacteria, the RNA is translated, producing the encoded product, such as, for example, an antigen.

Bacteria are advantageous vessels for RNA delivery, because bacteria readily can be cultured in large amounts, and then stored and/or formulated or provided as powder, a tablet, or a liquid for injection. They can be administered by any suitable route, including intravenously, and mucosally through the nose or lungs. They are very stable, and will provide large, stable quantities of RNA.

The immunostimulatory bacteria provided herein, when used for the delivery of RNA, such as mRNA, for example, when used as vaccines, or to encode viral antigens and/or other proteins to protect against pathogens, such as Severe Acute Respiratory Syndrome coronavirus 2 (SARS-CoV-2, which causes COVID- 19), can encode the full length wild-type SARS-CoV-2 spike protein (see, e.g., SEQ ID NO:438), or can encode the full length receptor binding domain (RBD) of the SARS- CoV-2 spike protein (see, SEQ ID NO:440), or can encode a portion of the spike protein RBD (see, e.g., SEQ ID NO:441) that is sufficient to induce or cause an immune response, and/or to immunize or vaccinate or protect a subject against SARS- CoV-2. The immunostimulatory bacteria also can deliver mRNA encoding mutants of the SARS-CoV-2 spike protein, with mutations in the spike protein or a portion thereof, such as the RBD or a portion thereof, that increase or enhance the expression of the spike protein or RBD or portions thereof, and/or that increase or enhance the binding of the spike protein or RBD, or the portions thereof to the ACE2 receptor.

The spike protein or spike protein RBD mutants include any described herein and known in the art, including, but not limited to, those comprising the mutations V367F, D614G, G476S, V483A, H49Y, N501Y, N501F, N501W, N501V, F817P, A892P, A899P, A942P, K986P, V987P, V417K, G502D, N501T, Q498Y, W436R and D364Y, and those comprising mutations at residues corresponding to residues N439/R426, L452/R426, T470/N457, E484/P470, Q498/Y484 and N501/T487 of the spike protein or portion thereof. The immunostimulatory bacteria can be used as vaccines against SARS-CoV-2 by delivering mRNA encoding any of the antigenic sequences or modified forms of the SARS-CoV-2 spike protein or RBD, including portions thereof, as described, for example, in U.S. Patent Nos. 10,973,908 and 10,702,600. The immunostimulatory bacteria also can be used as vaccines/for mRNA delivery, to deliver the mRNA in the Pfizer-BioNTech COVID-19 vaccine, and in the Modema COVID-19 vaccine, as described elsewhere herein. SARS-CoV-2 proteins for use as antigens, include, for example, spike protein, nucleocapsid and M protein and antigenic portions thereof. The immunostimulatory bacteria can be used to deliver universal flu antigens, such as flu vaccines that elicit antibodies against HA proteins from a variety of flu strains, such as influenza type A strains (Hl and H3) and influenza type B strains. Other combinations include repeating patters of HA epitopes, such as the FlusMos-vl. See, e.g., Nachbaguer et al. (2021) Nature Medicine 27: 106- 114, which provides a chimeric hemagglutinin-based universal influenza vaccine.

EXAMPLE 37

Immunomodulatory Strains Containing Human IL-15R«-IL-15sc and huSTING N154S/R284G tazCTT Colonize Autochthonous Murine Tumors

It previously has been reported that the attenuated Salmonella strain YS1646 is able to colonize tumors in the 4T1 murine breast transplant model (average tumor size 400 mm ³ , average CFUs/g = 10 ⁸ ), but is significantly less able to colonize the spontaneous BALB-NeuT breast tumor model (average tumor size 400 mm ³ , average CFUs/g = 10 ³ ; see, e.g., Drees et al. (2015) J. Cancer 6(9):843-848).

In order to show that the immunostimulatory bacteria provided herein do not have a similar deficiency in colonizing spontaneous tumors compared to transplanted tumors, tumors were collected from the orthotopically-transplanted EMT6 mouse model of breast cancer, and compared to tumors collected from the spontaneous MMTV-PyMT breast cancer model. For the EMT6 model, 6-8 week-old female BALB/c mice (5 mice per group) were inoculated orthotopically in the 4 ^th mammary fat pad with EMT6 cells (IxlO ⁶ cells in 100 pL PBS). Mice bearing 7-day established flank tumors (average tumor size of 56 mm ³ ) were IV injected with a single dose of 3 xlO ⁷ CFUs of the Y S I 646 Sasd! AFLG/ NpagP 1 Sansli I Scsgl) strain, containing a plasmid encoding the combination of huIL-15Ra-IL-15sc + huSTING N154S/R284G tazCTT. At day 4 post-IV dosing, mice were euthanized, and tumors were homogenized and plated on LB plates to enumerate the number of colony forming units (CFUs) per gram of tumor tissue. The

Y S I 646 Sasd! AFLG/ NpagP / NansB I Scsgl) strain colonized tumors at a mean of 1.7 xlO ⁷ CFUs per gram of tumor tissue.

For the spontaneous model of breast cancer, MMTV-PyMT mice were IV injected with 3 xlO ⁷ CFUs of the same bacterial strain when their largest tumors measured an average of 272 mm ³ in volume. Tumors were collected at day 6 post-IV dosing. Tumors were homogenized and plated on LB plates to enumerate the number of CFUs per gram of tumor tissue. All tumors collected were found to be well colonized despite a range of tumor weights (0.05 g to 0.36 g, N=7), with a mean of 2.89 x 10 ⁶ CFUs/g. These data demonstrate that, unlike the parental YS1646 strain, the immunomodulatory YS1646 Sa d! AFLG/ pagPI ansBI ScsgD strain, containing the plasmid encoding the combination of huIL-15Ra-IL-15sc + huSTING N154S/R284G tazCTT, was able to colonize spontaneous MMTV-PyMT tumors at a comparable level to the transplanted EMT6 tumors (P = 0.18, NS).

Since, unlike the transplanted tumors, the spontaneous tumors have vasculature that is more comparable to human tumors, these data indicate that the immunostimulatory bacteria will colonize human tumors at a much higher rate/level than the parental YS1646 strain was reported to in a phase I clinical trial of advanced cancer patients (see, e.g., Toso etal. (2002) J. Clin. Oncol. 20(1): 142-152).

EXAMPLE 38

Immunomodulatory Bacterial Strains, Containing Plasmids Encoding Human IL-15Ra-IL-15sc and huSTING N154S/R284G tazCTT, or a NanoLuciferase® Luciferase Control Plasmid, are Well Tolerated in Non-Human Primates

In a phase I human clinical trial, it was reported that the maximum tolerated dose (MTD) of the parental YS1646 strain is 3 x 10 ⁸ CFUs/m ² , and that the doselimiting toxicity (DLT) dose was 1 x 10 ⁹ CFUs/m ² (see, e.g., Toso et al. (2002) J. Clin. Oncol. 20(l):142-152). At these doses, the toxicities and adverse events, such as fever, hypotension, thrombocytopenia, anemia, vomiting, diarrhea, nausea, and hypophosphatemia, were attributed to very high serum levels of pro-inflammatory cytokines, measured at 4 hours post-IV dosing, including TNF-ot (approximately 500,000 pg/mL), IL-6 (approximately 500,000 pg/mL), and IL-ip (approximately 200 pg/mL). In a separate study, strain YS1646 was evaluated in a non-human primate (NHP) study using cynomolgus monkeys. In this study, 1 x 10 ⁹ CFUs/monkey was found to be the MTD (Human Equivalent Dose (HED) = 4 x 10 ⁹ CFUs/m ² ), while a dose of 1 x 10 ¹⁰ CFUs/monkey was considered not tolerated (HED = 4 x 10 ¹⁰ CFUs/m ² ). DLTs at the top dose were attributed to liver-associated adverse events (AEs), and serum cytokines were not measured (see, e.g., Lee etal. (2000) International Journal of Toxicology 19:19-25). Since the NHPs tolerated strain YS1646 much better than the humans, with an MTD value of a log higher than the human MTD, it can be inferred that the cytokine levels of the monkeys would be approximately a log lower than those measured in humans.

To determine the MTD and serum cytokine profile of immunostimulatory bacteria provided herein, such as strains that lack flagella and that are msbBIpagP', in NHPs, a tolerability study was performed. For this study, 15 previously untreated male cynomolgus monkeys, with ages ranging from 24 to 50 months, were utilized. The NHPs were IV administered the YS1646Aasz// AFLG/ kpagP! \ansH! ' kcsgD ft - ftpurl strain (discussed above), containing a plasmid encoding the combination of huIL-15Ra-IL-15sc + huSTINGN154S/R284G tazCTT, at doses of 3 x 10 ⁸ CFUs/monkey, 1 x 10 ⁹ CFUs/monkey, or 3 x 10 ⁹ CFUs/monkey (3 NHPs per dosing group), or were IV administered the YS 1646Aasr7/AFLG/ApagP/Aan&B/AcsgZ) strain, containing a plasmid encoding NanoLuciferase® luciferase, at a dose of 3 x 10 ⁸ CFUs/monkey (3 NHPs per group), and compared to the saline vehicle control group (N=3). The NHPs were bled prior to dosing, 4 hours post-dosing, and 24 hours postdosing, and serum cytokine levels were measured using a monkey cytokine U-Plex® panel (Meso Scale Discovery), according to the manufacturer’s protocol.

The results, which are summarized in the tables below, show that the bacterial strains were well-tolerated at all dose levels tested, and no clinical findings were reported that differed significantly from the saline (PBS) control group. Thus, no MTD could be established in this study. Serum cytokine levels were very low overall, particularly for the cytokines that were attributed to dose-limiting toxi cities (DLTs) in the human clinical trial with strain YS1646, such as the cytokines measured 4 hours post-IV dosing in the 3 x 10 ⁹ CFUs/monkey dose group (HED = 1.2 x IO ¹⁰ CFUs/m ² ), including TNF-a (mean concentration of 4.6 pg/mL), IL-6 (mean concentration of 376.9 pg/mL), and IL-lp (mean concentration of 0.88 pg/mL). Only the serum cytokine levels of IP- 10/C XCL 10 (mean concentration of 10,549.6 pg/mL) and MCP- 1/CCL2 (mean concentration of 6247.7 pg/mL) were high in the top dose level group, 4 hours post-dosing, and remained elevated at 24 hours post-dosing. These analytes are not associated with toxicity, and can indicate a more favorable immune profile. In summary, the immunostimulatory bacteria provided herein, including those containing plasmids encoding payloads, are very well-tolerated in NHPs. It can thus be inferred, from these data, that there will be a high level of tolerability of the immunostimulatory bacterial strains in humans. Serum Cytokines Levels Prior to Dosing

SD = standard deviation

Serum Cytokine Levels, 4 Hours After Dosing

SD = standard deviation

Serum Cytokine Levels, 24 Hours After Dosing

SD = standard deviation

EXAMPLE 39

Thymidine Auxotroph strains

The Salmonella typhimurium essential gene thyA encodes thymidylate synthase, a key enzyme in DNA synthesis. Deletion or mutation of the gene renders the strain auxotrophic for thymidine. Upon deprivation of the complementing substrate, bacterial cell death occurs and it has been demonstrated that no macromolecules are released from thyA mutants upon thymidine starvation (Loessner et al., FEMS Microbiol Lett. 265(1): 81-88 (2006)), reducing potential release of PAMPs and subsequent activation of innate inflammatory responses. To generate a thymidine auxotroph that is incapable of replication in vivo and rapidly dies in the absence of supplemented thymidine, thyA was deleted from strain YS1646 asd/AFLG/ApagP/AansB/AcsgD/F-ApurI.

A 690 bp region containing the thyA gene (SEQ ID NO: 464; DC51_3078), was targeted for deletion using modifications of the method of Datsenko and Wanner (Proc. Natl. Acad. Sci. U.S.A. 97:6640-6645 (2000)). To disrupt thyA gene, a thyA gene knockout cassette was constructed by overlapping PCR. This cassette contains a Kanamycin resistance (Kan ^R ) cassette flanked by two I-Scel cut site, and homologous regions on both the 5’ end (345 bp) and 3’ end (332 bp) to facilitate recombination. Immediately 5’ of the Kan ^R cassette there is 75 bp of DNA sequence that is perfectly homologous to 75 bp on the 3’ of the Kan ^R cassette. This repeat element (thyA MHA) leaves a scarless deletion upon excision of the Kan ^R cassette.

Specifically, with reference to the sequences in the table below, the thyA left homology arm sequence was amplified from YS1646 using primers thya-1 and thya- 2, was PCR overlapped with a Kan ^R cassette using primers thya-1 and scF \-thyA right homology arm sequence amplified from YS1646 using primer thya-3 and thya-4, was PCR overlapped with thyA MHA (prepared as a synthetic gBlock gene fragment from a supplier (IDT) using primer loxp-4 and thya-4).

Table 3: thyA gene KO homology arm length information

Table 4: thyA KO primer sequence information

>thyA MHA gBlock fragment

Note: thyA KO middle homology arm

(ggcttcccaaccttaccagagggcgccccagctggcaattccgGGTACCggcaagT AGGGATAACAGGGTAA TgttaTCTGCAgaccaaggacccagattatgcagcaacacgtttcctgaggaaccatgaa acagtatttagaactgatgcaaaaa gtgctggatgaaggcacacagtgaaattgaaggctatgatccgcac, SEQ ID NO: 479) was used for the lambda red recombination step during Kan gene removal using I-Scel.

The full length thyA gene knockout cassette then was constructed by overlap

PCR using primers thya-1 and thya-4 (see above Table), gel purified, and introduced into strain YS1646Aasd/AFLG/ApagP/AansB/AcsgD/F-ApurI carrying the temperature sensitive lambda red recombination plasmid pSL0304 by electroporation. The kanamycin resistance gene then was cured by I-Scel/lambda red mediated recombination as previously described Yang and Yang (Applied and Environmental Microbiology, 80: 3826-3834 (2014)), and the temperature-sensitive plasmids were cured by growth at non-permissive temperature. The thyA gene fragment knockout sequence was confirmed by PCR using primers thya-5 and thya-6 (Table 2), and verified by DNA sequencing. The resulting mutant derivative of parental strain, Y S I 646J<7.SZ// EFEG/ApagP, AansB, Acsgl) E-Apurl, was designated YS 16A6Aasdl EFEGIApagP/AansB/AcsgD/F-Apurl/A thyA .

Thymidine deletion strains were grown in a DasGip fermentation system supplemented with 250/zg/mL thymidine for 6 hours and culture material was subsequently processed for injection stocks. NS\6A6AasdllAEEGIApagP/AansB/AcsgD/F-ApurI/AthyA was cultured in the presence of 250/zg/mL thymidine (Sigma Aldrich T1895-1G) and 50/zg/mL diaminopimelic acid to generate electrocompetent cells.

\E\6A6Aasd! MLGIApagP AansB, Acsgl) F-Apurl AthyA was transformed with plasmids designated ADN-86, ADN-870 or ADN-872, to generate strains STST-321, STST-326 and STST-328 (see table below), respectively.

Y S I 646J«.sz// EPEGIApagP/AansB/AcsgD/F-Apurl/AthyA strains, designated STST- 321, STST-326 and STST-328 containing the noted plasmids (see Table below), were grown in a DasGip fermentation system at IL scale for 6hrs in the presence of 250jUg/mL thymidine in parallel with CRST-2000,

Y S I 646 AW/ MLGIApagP AansB, Acsgl) F-Apurl (resulting strain designated test strain 4, for example in the table below) bearing a plasmid comprising the sequence set forth in SEQ ID NO: 501, tested in the absence of thymidine supplementation.

The strain referred to as “CRST-2000,” includes a plasmid that comprises nucleic acid having the sequence set forth in SEQ ID NO:501. SEQ ID NO:501 sets forth the sequence of a construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a VCIP IRES (nucleotides 2256-2802); human IL-15 complex (in which nucleotides 2803-3126 are human IL-15Ra leader + sushi, nucleotides 3127-3186 are a (Gly4- Ser)4 linker, and nucleotides 3187-3531 are human IL-15 extracellular); a T2A sequence (nucleotides 3541-3594); human STING with the N154S/R284G mutations and a Tasmanian devil CTT (nucleotides 3595-4731); an HPRE (nucleotides 4741- 5274); a bGH poly(A) signal (nucleotides 5283-5507); and asd (nucleotides 5858- 6297 and 1-736). SEQ ID NOs:546 and 547 set forth the nucleotide sequence of the STING N154S/R284G Tasmanian devil CTT and human IL-15 complex, respectively. The amino acid sequence of a human STING protein with the N154S/R284G mutations and a Tasmanian devil CTT is set forth in SEQ ID NO: 500. The amino acid sequence of the human IL-15 complex in the construct is set forth in SEQ ID NO 548.

Fermentation culture was analyzed for ODeoo throughout the fermentation. The growth of the thymidine deletion strains were observed to be comparable to the CRST-2000 control. To confirm thymidine auxotrophy, growth of strains STST-321, STST-326 and STST-328 was evaluated in medium with or without 250 _j ug/mL thymidine supplementation in comparison to control CRST-2000. Injection stocks were thawed at room temperature and 2.5/xL was used to inoculate 250/zL media +/- thymidine in triplicate wells in a clear bottom 96-well plate; growth was evaluated for 12 hours, shaking at 37°C, by OD600 in a SpectraMax® microplate reader (Molecular Devices). Thymidine deletions strains were grown in broth with or without 250 ₍ ug/mL thymidine supplementation, and ODgoo was monitored in 15 minutes intervals over the course of 12 hours.

The results show that no growth was observed for STST-321, STST-326 and STST-328 in the absence of thymidine supplement, while CRST-2000 demonstrated robust growth to stationary phase. In the presence of 250/rg/mL thymidine supplementation, STST-321, STST-326 and STST-328 grew to stationary phase with a growth profile similar to CRST-2000, demonstrating precise and controlled auxotrophy for thymidine.

Bacterial strains, such as these strains, have a variety of advantageous effects, detailed throughout the disclosure herein, upon administration. See, for example, Figure 17F, which shows that immunostimulatory bacteria provided herein act as a comprehensive immunotherapy. EXAMPLE 40

Optimized Expression of IRF3, IFNa2, and IFN-0 Expression Cassettes

Plasmids containing combinations of human IRF3 S396D or mouse IRF3 S388D, or human or mouse IFNa2, or human or mouse IFN-P, where the bi-cistronic constructs contain T2A peptides, were tested for expression of each encoded payload/product against single expresser controls.

HEK293T STING Null Cells (293 -Dual™ Null Cells; InvivoGen), which do not contain endogenous STING, and express secreted embryonic alkaline phosphatase (SEAP), placed under the control of the endogenous IFN-stimulated response element (ISRE) promoter, where the coding sequence of ISRE is replaced by the SEAP ORF using knock-in technology, were used. STING activity, thus, is assessed by monitoring type I IFN induced SEAP production. The 293-Dual™ Null cells also express Lucia™ luciferase, a secreted luciferase, placed under the control of the endogenous IFN-P promoter; the coding sequence of IFN-P has been replaced by the Lucia™ luciferase ORF using knock-in technology. This allows for the assessment of STING activity by monitoring the expression of IFN-p. Using these cells, STING activity can be assessed by monitoring ISRE-induced SEAP production and/or IFN- -dependent expression of Lucia™ luciferase. The two reporter proteins, SEAP and Lucia™ Luciferase, can be measured in the cell supernatant using standard assays and detection reagents, such as the QUANTLBlue™ and QUANTI-Luc™ detection reagents (InvivoGen), respectively.

The cells were seeded in 24-well plates coated with poly-L-lysine at 200,000 cells per well, and incubated overnight at 37 °C in a 5% CO2 incubator, to achieve 80% confluency. The following day, 300 ng of each plasmid DNA, and 40 ng of a CMV-GFP vector (/.<?., a vector encoding green fluorescent protein under the control of a CMV promoter), were diluted in serum-free media and added to FuGENE® transfection reagent (Promega), at the proper reagent:DNA ratios, with untransfected wells serving as negative controls (in duplicates). Cell culture supernatants from each sample were collected 48 hours post-transfection.

The activity of the constructs were evaluated using the ISRE-SEAP and IFN- -Lucia™ reporter systems. The type I interferon (IFN) activity was assessed by monitoring type I IFN-stimulated SEAP production in the cell supernatants. 20 pL of cell culture supernatant was added to 180 pL of QUANTLBlue™ reagent (InvivoGen), which is used for measuring SEAP. Type I interferon activation was determined by measuring ISRE-induced SEAP activity on a SpectraMax® M3 Spectrophotometer (Molecular Devices), at an absorbance wavelength of 650 nm. The type I interferon (IFN) activity also was assessed by monitoring the type I IFN- stimulated Lucia™ luciferase production in the cell supernatants. 20 pL of cell culture supernatant was added to 50 pL of QUANTI-Luc™ reagent (InvivoGen), which is used for measuring Lucia™ luciferase activity. Type I interferon activation was determined by measuring IFNp-induced Lucia™ luciferase activity on a SpectraMax® M3 Spectrophotometer (Molecular Devices), on the luminescence setting.

A. Cytokines Resulting from Transfection of Mouse IRF3, IFNa2, and IFNb in BMDCs

Various single expressing plasmids and combination plasmids containing mouse IRF3 S388D, mouse IFNa2, mouse IFN-0 were compared to the hSTING N154S/R284G tazCTT plasmid. The expression results are provided in the table below. ISRE-SEAP reporter activity is higher than hSTING N154S/R284G tazCTT in all of the human constructs, as well as mIFNa2 T2A mIRF3 S388D and mIRF3 S388D T2A mIFNa2. IFN -Lucia reporter activity is higher than hSTING N154S/R284G tazCTT for the phosphomimetic hIRF3 alone, hIFNa2 T2A hIRF3 S388D and hIRF3 S388D T2A hIFNa2. These results demonstrate high levels of type I IFN production can be induced through expressing individual type I IFN signaling components, as well as combinations of constitutive type I IFN signaling components. Mouse and Human IRF3, IFNa2, and IFN-0 Expression Normalized by Transfection Efficiency

- 490 -

RECTIFIED SHEET (RULE 91 ) ISA/EP

B. Plasmids containing combinations of human IRF3 S396D or mouse IRF3 S388D, or human or mouse IFNoc2, or human or mouse IFN-P, and combinations thereof, where the bi-cistronic constructs contain T2A peptides, were tested for expression by transfection in murine primary bone marrow-derived dendritic cells (BMDCs).

To test these, Golden Ticket murine bone marrow was isolated and flushed into 1.5 mL Eppendorf tubes, and spun at 1200 RPM for 5 minutes, to collect the bone marrow cells. Cells were washed once in RPMI-1640 + 10% FBS, then seeded in 96-well TC-treated plates in RPMI-1640 + 10% FBS with 20 ng/ml GM-CSF After four days, non-adherent cells were pipetted off the wells and re-seeded at 2e5 cells per well in RPMI-1640 + 10% FBS in a 96-well plate for transfection. Cells were transfected using Viromer® RED, according to the manufacturer’s instructions. Briefly, 300 ng of plasmid DNA, as well as a “No DNA” control, were diluted in the provided buffer, and mixed with 0.08 pL of Viromer® RED and incubated at room temperature for 15 minutes to allow the DNA/Viromer® RED complexes to form. The DNA/Viromer® RED complexes were then slowly added to each well of the 96- well plate (in duplicates), and the plate was incubated at 37 °C in a CO2 incubator. Supernatants were harvested at 48 hours, and assayed for murine IFNcc, IFN-P, CXCL10 (IP-10), and IL-6 using a flow cytometry-based cytokine bead array (CBA), according to the manufacturer’s protocol.

Various single expressing plasmids and combination plasmids encoding mouse IRF3 S388D, mouse IFNoc2, mouse IFN-p were compared to the hSTING N154S/R284G tazCTT plasmid. As shown in the table below, transfections with constructs mIFNoc2, mIFN-P, and mIFNoc2 T2A mIFN-P had higher levels of IFN-oc and IFN-p than hSTING N154S/R284G tazCTT. mIFN-p and mIFNa2 T2A mIFN-p constructs had higher levels of CXCL10 than hSTING N154S/R284G tazCTT. mIRF3 S388D, mIFNoc2 T2A mIRF3 S388D, and mIRF3 S388D T2A mIFNa2 constructs all had lower IL-6 levels than hSTING N154S/R284G tazCTT. These data demonstrate the ability to induce type I IFN signaling through multiple combinations of expressed type I IFN components, alone and in combination, and with minimal induction of pro-inflammatory IL-6. Expression of IFN-P alone induces the highest type I IFN activity of all the targets tested.

Cytokines Resulting from Transfection of Mouse IRF3, IFN«2, and IFN-P in BMDCs

C. Mouse IFNa2 T2A IFN-P is Expressed in vivo in a Murine Orthotopic Breast Cancer Model

For this experiment, 6-8 week-old female BALB/c mice (5 mice per group) were inoculated in the left mammary fat pad with EMT6 tumor cells (ATCC # CRL- 2755) (IxlO ⁶ cells in 100 pL PBS). Mice bearing 9 day-old established mammary tumors (-100 mm ³ in volume) were IV injected with a single dose of 2 xlO ⁷ CFUs of the YS1646 AaxiZ/AFLG/A^agP/AaHxB/AcxgD/A/JMr/Qarge clean) strain containing a bicistronic plasmid encoding mIFNa2 T2A mIFN-P, and compared to PBS control.

The level of tumor mIFNa2 and mIFN-P expression, relative to actin, was determined in order to assess tumor-specific payload delivery. On day 4 post-FV injection, tumors were excised and processed for RNA extraction using the TissueLyser II (Qiagen) in 1.2 mL of RLT plus lysis buffer. The homogenates were collected for RNA isolation using the RNeasy® Plus Mini Kit (Qiagen), according to the manufacturer’s instructions. Total RNA concentration was measured using a NanoDrop™ 2000 UV-Vis Spectrophotometer (Thermo Scientific) The purity of each sample also was assessed from the A260/A230 absorption ratio. Synthesis of cDNA was performed from 0.5-1 pg of template RNA using a CFX96™ Real-Time System (Bio-Rad) and i Script™ gDNA Clear cDNA Synthesis Kit (Bio-Rad) in a 20 pL reaction, according to the manufacturer’s instructions. qPCR was performed with a CFX96 Real-Time System (Bio-Rad). PrimePCR™ Probe Assay for mfFNot.2 (qMmuCEP0043629), mIFNpl (qMmuCEP0058870) were purchased from Bio-Rad. The qPCR reaction (20 pL) was conducted per protocol, using either the SsoAdvanced Universal SYBR® Green Supermix or the iQ Multiplex Powermix (Bio-Rad). The standard thermocycling program on the Bio-Rad CFX96 Real-Time System consisted of a 95 °C denaturation for 150 sec, followed by 39 cycles of 95 °C for 15 sec and 60 °C for 55 sec. Quantification of the target mRNA was normalized using Actin reference mRNA (Bio-Rad, qMmuCEP0039589). AC _q was calculated as the difference between the target and reference gene.

The values are shown in the table below, as the average of five mice (PBS group) or three mice (mIFNa2 + mIFN-P group). Compared to the PBS control group, ¥S1646Arzs’6?/AFLG/AprzgF ^, /Arzt5S/Ac5gD/A/>Mr/(iargeciean) strain containing plasmids encoding mIFNa2 and mIFN-P significantly increased expression of mIFNa2 and mIFN-P genes in EMT6 tumors.

In order to measure plasmid delivery to the tumor, and subsequent heterologous gene expression and protein secretion, the amount of mIFNa2 and mIFN-P protein was measured in these tumors. For this, lysates were harvested from homogenized tumors and assessed for protein expression using a flow cytometrybased cytokine bead array (CBA), according to the manufacturer’s protocol.

As shown in the table below, the YS1646Arzs’6?/AFLG/AprzgF ^, /A«t55/Ac5gD/A/?Mr/(iargeciean) strain containing plasmids encoding mIFNa2 and mIFN-P demonstrated high levels of mIFNa2 and mIFN-P protein per gram of tumor, as compared to the PBS control tumors, demonstrating potent delivery of mlFNoiZ and mIFN-P into tumors following IV administration.

Additional constructs encoding type I interferons (IFNs) are described in Example 57 below. Those constructs and plasmids containing them can be introduced into the immunostimulatory bacteria provided herein, such as the strain YS1646Aa%// AFLG/ SpagP! kansB! KcsgD/kpurI(i _arge dean) and related strains such as those strains that lack flagella (AFLG) are csglX and msbBr/pcigP~ described throughout the disclosure herein.

EXAMPLE 41

Expression of Membrane Bound Human and Mouse IL-12 in Mouse and Human Cells

Mouse membrane bound IL-12p70 (SEQ ID NO:466) was constructed with mouse IL-12p40 (residues 3009-4010 of SEQ ID NO:399) followed by a 15 amino acid linker (SEQ ID NO:480; GGGGSGGGGSGGGGS), and IL-12p35 (residues 4056-4634 of SEQ ID NO:399). The transmembrane and cytosolic portions of murine CD80 (SEQ ID NO:481;

PPEDPPDSKNTLVLFGAGFGAVITVVVIVVIIKCFCKHRSCFRRNEASRETNNS LTFGPEEALAEQTVFL) follow this sequence. Human membrane bound IL-12p70 (SEQ ID NO:467) was constructed with human IL-12p40 (SEQ ID NO: 482;

MCHQQLVISWFSLVFLASPLVAIWELKKDVYVVELDWYPDAPGEMVVLTCD TPEEDGITWTLDQSSEVLGSGKTLTIQVKEFGDAGQYTCHKGGEVLSHSLLLL HKKEDGIWSTDILKDQKEPKNKTFLRCEAKNYSGRFTCWWLTTISTDLTFSV KSSRGSSDPQGVTCGAATLSAERVRGDNKEYEYSVECQEDSACPAAEESLPIE VMVDAVHKLKYENYTSSFFIRDIIKPDPPKNLQLKPLKNSRQVEVSWEYPDT WSTPHSYFSLTFCVQVQGKSKREKKDRVFTDKTSATVICRKNASISVRAQDR YYSSSWSEWASVPCS) followed by a 15 amino acid linker sequence (GGGGSGGGGSGGGGS, SEQ ID NO: 480) and human IL-12p35 (RNLPVATPDPGMFPCLHHSQNLLRAVSNMLQKARQTLEFYPCTSEEIDHEDI TKDKTSTVEACLPLELTKNESCLNSRETSFITNGSCLASRKTSFMMALCLSSIY EDLKMYQVEFKTMNAKLLMDPKRQIFLDQNMLAVIDELMQALNFNSETVPQ KSSLEEPDFYKTKIKLCILLHAFRIRAVTIDRVMSYLNAS, SEQ ID NO:483). The transmembrane and cytosolic portions of human CD80 (SEQ ID NO:484; DNLLPSWAITLISVNGIFVICCLTYCFAPRCRERRRNERLRRESVRPV) follow this sequence.

HEK cells were seeded in 24-well plates coated with poly-L-lysine at 200,000 cells per well, and incubated overnight at 37 °C in a 5% CO2 incubator, to achieve 80% confluency. The following day, 300 ng of each plasmid DNA, and 40 ng of a CMV-GFP vector (i.e., a vector encoding green fluorescent protein under the control of a CMV promoter), were diluted in serum-free media and added to FuGENE® transfection reagent (Promega), at the proper reagent:DNA ratio, with untransfected wells serving as negative controls (in duplicates). Cell culture supernatants from each sample were collected 48 hours post-transfection.

The activity of the constructs were evaluated using a co-culture of transfected HEK cells with HEK-Blue IL-12 (InvivoGen) cells. 24 hours post transfection, 5e4 HEK cells were detached and seeded in a 96-well TC plate. Additionally, 5e4 HEK- Blue IL-12 cells were seeded in the same wells. 48 hours post transfection, cell culture supernatants from each sample were collected 48 hours post-transfection. Human and mouse IL- 12 bioactivity was determined by measuring the SEAP reporter, which measures activation of the STAT -4 pathway downstream of IL-12 binding to its receptor. IL- 12 bioactivity was determined by measuring ISRE-induced SEAP activity on a SpectraMax® M3 Spectrophotometer (Molecular Devices), at an absorbance wavelength of 650 nm.

The expression results of the constructs are provided in the table below. Each of the murine and human membrane bound IL-12p70 activated the STAT4-SEAP reporter in HEK-Blue IL-12 cells to at least the same extent as soluble murine and human IL-12p70, thereby confirming the functionality of constructs.

Mouse and Human Membrane Bound IL-12p70 Expression Normalized by Transfection Efficiency

Membrane bound IL-12p70 constructs were tested for expression on the surface of transfected HEK cells by flow cytometry. HEK wildtype cells transfected with murine or human IL-12p70, using the same conditions as above, and were harvested 48 hours post transfection. Cells were detached by PBS and seeded in wells of a V-bottom 96-well plate. The cells were washed once with PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes. The cells were then resuspended in 50 uL PBS + 2% FBS containing either anti-human IL-12p70 APC (Miltenyi) diluted 1 :10 or anti-mouse IL-12p70 APC (Miltenyi) diluted 1:10, and incubated on ice for 30 minutes in the dark. The cells were washed twice with PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes and resuspended in DAPI diluted 1:6000 in PBS + 2% FBS. Flow cytometry data were acquired using the ACEA NovoCyte® flow cytometer (ACEA Biosciences, Inc.), and analyzed using the FlowJo™ software (Tree Star, Inc.). Only 1.67% of untransfected cells were APC positive, while the membrane bound human IL-12p70 was 40.4% positive and the membrane bound murine IL-12p70 was 59.9%. Therefore, these membrane bound IL-12 constructs are well expressed on human cells.

Plasmids containing human or murine membrane bound IL-12p70 were tested for expression by transfection in murine primary bone marrow-derived dendritic cells (BMDCs). Wild type murine bone marrow was isolated and flushed into 1.5 m Eppendorf tubes, and spun at 1200 RPM for 5 minutes, to collect the bone marrow cells. Cells were washed once in RPML1640 + 10% FBS, then seeded in 96-well TC- treated plates in RPMI-1640 + 10% FBS with 20 ng/ml GM-CSF. On Day 3, additional RPMI-1640 + 10% FBS with 20 ng/ml GM-CSF was added to the cells. After six days, non-adherent cells were pipetted off the wells and re-seeded at 2e5 cells per well in RPMI-1640 + 10% FBS in a 96-well plate for transfection. Cells were transfected using Viromer® RED transfection reagent, according to the manufacturer’s instructions. Briefly, 300 ng of plasmid DNA, as well as a “No DNA” (transfection reagent only) control, were diluted in the provided buffer, and mixed with 0.08 pL of Viromer® RED transfection reagent and incubated at room temperature for 15 minutes to allow the DNA/Viromer® RED complexes to form. The DNA/Viromer® RED complexes were then slowly added to each well of the 96- well plate (in duplicates), and the plate was incubated at 37 °C in a CO2 incubator. Cells were harvested 48 hours post transfection and were assayed for IL- 12 surface expression using flow cytometry.

The cells were washed once with PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes. The cells were then resuspended in 50 pL PBS + 2% FBS containing either anti-human IL-12p70 APC (Miltenyi) diluted 1:10 or anti-mouse IL- 12p70 APC (Miltenyi) diluted 1 : 10, CDl lb PE diluted 1:200, ClassII APC-Cy7 diluted 1 :200, and incubated on ice for 30 minutes in the dark. The cells were washed twice with PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes and resuspended in DAPI stain diluted 1 :6000 in PBS + 2% FBS. Flow cytometry data were acquired using the ACEA NovoCyte® flow cytometer (ACEA Biosciences, Inc ), and analyzed using the Flowlo™ software (Tree Star, Inc.). In the dendritic cell population, only 0.205% ± 0.021% of untransfected cells and 0.685% ± 0.262% of VCIP IRES CMVNanoLuc® transfected cells were APC positive while the membrane bound human IL-12p70 was 6.14% ± 1.27% positive and the membrane bound murine IL-12p70 was 2.64% ± 0.87% positive. These results show that the membrane bound IL- 12 constructs were detected on the surface of mouse bone marrow dendritic cells above background.

Plasmids containing murine membrane bound IL-12p70 also were tested for expression by transfection in murine primary bone marrow-derived macrophages (BMMs). Wild type murine bone marrow was isolated and flushed into 1.5 ml Eppendorf tubes, and spun at 1200 RPM for 5 minutes, to collect the bone marrow cells. Cells were washed once in RPMI-1640 + 10% FBS, then seeded in 24-well TC- treated plates in RPMI-1640 + 10% FBS with 20 ng/ml M-CSF. After three days, media was aspirated and replenished with 20 ng/ml M-CSF. Cells were transfected using Viromer® RED transfection reagent, according to the manufacturer’s instructions. Briefly, 750 ng of plasmid DNA, as well as a “No DNA” control, were diluted in the provided buffer, and mixed with Viromer® RED transfection reagent and incubated at room temperature for 15 minutes to allow the DNA/Viromer® RED complexes to form. The DNA/Viromer® RED complexes were then slowly added to each well of the 96-well plate (in duplicates), and the plate was incubated at 37 °C in a CO2 incubator. Cells were harvested 48 hours post transfection and were detached for IL- 12 surface expression using flow cytometry. The cells were detached with lOmM EDTA and washed once with PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes. The cells were then resuspended in 50 uL PBS + 2% FBS containing either anti-human IL-12p70 APC (Miltenyi) diluted 1 : 10 or anti-mouse IL-12p70 APC (Miltenyi) diluted 1 : 10, CD1 lb PE diluted 1 :200, ClassII APC-Cy7 diluted 1 :200, and incubated on ice for 30 minutes in the dark. The cells were washed twice with PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes and resuspended in DAPI diluted 1:6000 in PBS + 2% FBS. Flow cytometry data were acquired using the ACEA NovoCyte® flow cytometer (ACEA Biosciences, Inc ), and analyzed using the FlowJo™ software (Tree Star, Inc.). Only 0.01% ± 0.003% of untransfected cells and 0.36% ± 0.028% of VCIP IRES CMV NanoLuc® transfected cells were APC positive while the membrane bound murine IL-12p70 was 1.945% ± 0.106%. Therefore, these membrane bound IL-12 constructs were detected on the surface of mouse bone marrow macrophages above background.

EXAMPLE 42

Expression of DLL3xCD3 Bispecific T-cell Engagers from Human Cells

DLL3xCD3 bispecific T-cell engagers (sold under the trademark BiTE®), which targets delta-like ligand 3, which is selectively expressed in small-cell lung cancer, and CD3+ T cells, containing a FLAG tag, were designed using the amino acid sequences of anti-DLL3 antibodies SC16.15, SC16.34, and SC16.56, see e.g., International Patent Publication No. WO 2017/031458 A2. The anti-CD3 arm of this BiTE construct was designed from clone 145-2C11. These antibodies are fully murine sequences cloned into the pATI-1.76 vector under the control of a CMV promoter and with a 3’ HPRE. For human administration, the antibodies can be humanized as appropriate. The sequences were confirmed by Sanger sequencing.

The sequence of the resulting constructs are as follows:

SC16.56 DLL3HLxCD3HL (SEQ ID NO:485)

The construct includes, in the following order: mouse IgGK leader, SC16.56 VH, 15 amino acid GS linker, SC16.56 VL, 5 amino acid linker, 145-2C11 VH, 15 amino acid GS linker, 145-2C11 VL, and Flag tag, as set forth in SEQ ID NO: 485, and below:

SC16.56 DLL3LHxCD3HL (SEQ ID NO: 486)

The construct includes, in the following order: mouse IgGK leader, SC16.56

VL, 15 amino acid GS linker, SC16.56 VH, 5 amino acid linker, 145-2C11 VH, 15 amino acid GS linker, 145-2C11 VL, and Flag tag, as set forth in SEQ ID NO: 486, and below:

SC16.15 DLL3HLxCD3HL (SEQ ID NO 487) The construct includes, in the following order: mouse IgGK leader, SC16.15 VH, 15 amino acid GS linker, SC16.15 VL, 5 amino acid linker, 145-2C11 VH, 15 amino acid GS linker, 145-2C11 VL, and Flag tag, as set forth in SEQ ID NO: 487, and below:

SC16.15 DLL3LHxCD3HL (SEQ ID NO:488)

The construct includes, in the following order: mouse IgGK leader, SC16.15 VL, 15 amino acid GS linker, SC16.15 VH, 5 amino acid linker, 145-2C11 VH, 15 amino acid GS linker, 145-2C11 VL, and Flag tag, as set forth in SEQ ID NO: 488, and below:

SC16.34 DLL3HLxCD3HL (SEQ ID NO:489)

The construct includes, in the following order: mouse IgGK leader, SC16.34

VH, 15 amino acid GS linker, SC16.34 VL, 5 amino acid linker, 145-2C11 VH, 15 amino acid GS linker, 145-2C11 VL, and Flag tag, as set forth in SEQ ID NO: 489, and below:

SC16.34 DLL3LHxCD3HL (SEQ ID NO: 490)

The construct includes, in the following order: mouse IgGK leader, SC16.34

VL, 15 amino acid GS linker, SC16.34 VH, 5 amino acid linker, 145-2C11 VH, 15 amino acid GS linker, 145-2C11 VL, and Flag tag, as set forth in SEQ ID NO: 490, and below:

1.5 x 10 ⁶ HEK293T cells were plated one day prior on 6-well plates coated with poly-L-lysine, to achieve 80% confluency. On the day of transfection, 3 pg of DNA was diluted in serum-free media and added to FuGENE® transfection reagent (Promega), at the proper reagent:DNA ratios. Cell culture supernatants from each sample were collected after 48 hours of incubation. Some supernatant was concentrated in a 10 kDa spin column (Millipore).

The functionality of these cells transfected with the DLL3xCD3 BiTE encoding plasmids was demonstrated by binding of the DLL3xCD3 bispecific T-cell engagers on the cells to SHP77 cells (ATCC), which have DLL3 on the surface. 500,000 SHP77 cells were seeded in wells of a V-bottom 96-well plate. The cells were washed once with PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes. The cells were resuspended in 50 uL PBS + 2% FBS containing concentrated HEK supernatant corresponding to either untransfected cells or cells transfected with a BiTE. After 30 minutes, the cells were washed twice with PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes. The cells were then resuspended in 50 uL PBS + 2% FBS containing either anti -Flag biotin (Sigma) at 1 : 100 dilution or Protein L biotin (Genscript) at 1:200 dilution, and incubated on ice for 30 minutes. The cells were washed twice with PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes. Subsequently, the cells were stained with Streptavidin APC (Biolegend) at 1:200 dilution and incubated on ice for 30 minutes in the dark. The cells were washed twice with PBS + 2% FBS by centrifugation at 1300 RPM for 3 minutes and resuspended in DAPI diluted 1 :6000 in PBS + 2% FBS. Flow cytometry data were acquired using the ACEA NovoCyte® flow cytometer (ACEA Biosciences, Inc.), and analyzed using the FlowJo™ software (Tree Star, Inc.).

The table below provides the percent of positive cells gated as APC positive, corresponding to cells stained with BiTE and detected either by Protein L or anti-Flag antibody. SC16.56 DLL3LHxCD3HL, SC16.34 DLL3HLxCD3HL, and SC16.34 DLL3LHxCD3HL BiTEs all have similar detection on SHP77 cells while the other BiTEs are not detected. Therefore, SC16.56 DLL3LHxCD3HL, SC16.34 DLL3HLxCD3HL, and SC16.34 DLL3LHxCD3HL bind to DLL3.

Binding of cells expressing the DLL3xCD3 bispecific T-cell engagers DLL SHP77 cells

The expressed DLL3xCD3 bispecific T-cell engagers also were evaluated for binding to DLL3 and to CD3 with an ELISA. This ELISA was developed to detect correctly folded DLL3xCD3 bispecific T-cell engager binding to DLL3 and CD3 in the same assay. Human DLL3 (R&D Systems) was coated overnight at 4 °C on a high protein-binding 96-well plate, at a concentration of 1 ug/ml. The wells then were washed three times with PBS 0.05% Tween-20, and the wells were blocked with ELISA blocking buffer for 1 hour at room temperature. The wells then were washed three times with PBS 0.05% Tween-20. HEK293T cell culture supernatants, containing the DLL3xCD3 bispecific T-cell engagers, were added to the wells and incubated for 2 hours at room temperature. The wells were washed three times with PBS 0.05% Tween-20, and human IgG Fc conjugated CD3e was added to the wells (Aero Biosystems), and incubated for one hour at room temperature. The wells again were washed three times with PBS 0.05% Tween-20, and horseradish peroxidase (HRP)-conjugated anti-human antibody was added for a 1 hour incubation. The wells were then washed three times with PBS 0.05% Tween-20, and detection reagent (3,3',5,5'-tetramethylbenzidine (TMB), Thermo Fisher Scientific) was added to the wells. The enzymatic reaction was stopped with sulfuric acid (BioLegend), and the optical densities were read at 450 nm.

The table below provides the results of this ELISA, shown as the absorbance at 450 nm. All of the SC 16.56 DLL3LHxCD3HL, SC 16.34 DLL3HLxCD3HL, and SC 16.34 DLL3LHxCD3HL BiTE antibodies were detected, showing that these BiTEs antibodies are fully functional. SC16.34 DLL3LHxCD3HL has the highest binding in this ELISA, but since cell supernatants were used in this assay, it is not possible to distinguish whether this results from a higher concentration of this bispecific T-cell engager, or higher affinity to its targets compared to the other 3 bispecific T-cell engagers. In a separate experiment, the bispecific T-cell engagers were purified using ANTI-FLAG M2 Affinity Gel (Sigma) and run on this ELISA. SC 16.34 DLL3LHxCD3HL was the only purified BiTE to show ELISA signal above background.

BiTE binding by ELISA

EXAMPLE 43

Anti-CTLA-4 scFv-Fc Demonstrates Blockade of the CD80/CTLA-4 and CD86/CTLA-4 Interactions and Co-Expresses with human STING N154S/R284G tazCTT

An scFv-Fc specific for human CTLA-4 (see, SEQ ID NOs:427 and 428 for the nucleic acid and protein sequences, respectively) was designed using the amino acid sequence of ipilimumab. Ipilimumab is a fully human IgGlK monoclonal antibody that specifically binds human CTLA-4 (see, e.g., the antibody designated 10D1 in U.S. Patent Publication No. 2002/0086014 and in U.S. Patent No. 6,984,720), blocking CTLA-4’ s immune inhibiting interaction with CD80 (also known as B7.1 or B7-1) and CD86 (also known as B7.2 or B7-2).

To generate the ipilimumab scFv antibody fragment (see, SEQ ID NO:429), the variable light chain (VL) and variable heavy chain (VH) of ipilimumab were linked with a 20 amino acid long glycine-serine (GS) linker ((GGGGS)4). To generate the scFv-Fc antibody fragment (see, SEQ ID NO:428), the variable heavy chain of the ipilimumab scFv was linked to a human IgGl Fc, containing a mutation of the free cysteine in the hinge region to a serine (at position 272 in SEQ ID NO:428). The leader sequence (METPAQLLFLLLLWLPDTTG; corresponding to residues 1-20 in SEQ ID NO:428) was derived from the sequence of the human immunoglobulin kappa variable 3-20 (IGKV3-20) protein. The sequence was codon optimized for human using the GenScript GenSmart™ Codon Optimization tool (ATGGAGACACCTGCCCAGCTGCTGTTCCTGCTGCTGCTGTGGCTGCCCGA CACCACCGGC); the sequence is set forth in SEQ ID NO:491.

The neutralizing ability of the anti-CTLA-4 scFv-Fc was determined using competitive ELISAs to measure the ability of the antibody fragment to block the interactions between CTLA-4 and its ligands, CD80 and CD86. HEK293T cells were transfected with 3 micrograms of DNA encoding the anti-CTLA-4 scFv-Fc antibody fragment construct, using the FuGENE® transfection reagent (Promega), at the proper reagent:DNA ratios. Forty-eight hours post-transfection, HEK293T cell-free culture supernatants were harvested, filtered, and used in a competitive ELISA to assess the blockade activity of the anti-CTLA-4 antibody fragment.

For the competitive ELISAs, human CD80 or CD86 recombinant proteins (R&D Systems) were coated overnight at 4 °C on a high protein-binding 96-well plate, at a concentration of 100 ng/ml. The wells were then washed three times with PBS 0.05% Tween-20, and the wells were blocked with ELISA blocking buffer for 1 hour at room temperature. The wells were then washed three times with PBS 0.05% Tween-20. HEK293T cell culture supernatants, containing each of the anti-CTLA-4 antibody fragments, were mixed with 10 ng/ml of a recombinant human CTLA-4- human IgGl Fc chimera (R&D Systems), and added to the wells and incubated for 2 hours at room temperature. The wells were then washed three times with PBS 0.05% Tween-20, and horseradish peroxidase (HRP)-conjugated anti-human IgGl antibody was added to the wells (Jackson ImmunoResearch), and incubated for one hour at room temperature. The wells were then washed three times with PBS 0.05% Tween- 20, and detection reagent (3,3',5,5'-tetramethylbenzidine (TMB), Thermo Fisher Scientific) was added to the wells. The enzymatic reaction was stopped with sulfuric acid (BioLegend), and the optical densities were read at 450 nm.

The results of the competitive ELISAs are summarized in the table below. The anti-CTLA-4 scFv-Fc blocked the binding of CTLA-4 to CD86 by 94.14% and blocked the binding of CTLA-4 to CD80 by 83.46%. The CTLA4 scFv-Fc had a higher degree of CD86/CTLA-4 blocking activity compared with CD80/CTLA-4 blocking activity.

Competitive ELISA Results

Human anti-CTLA4 scFv-Fc (SEQ ID NO:427) was cloned into expression vectors with huSTING N154S/R284G tazCTT either containing or not containing the Vascular Endothelial Growth Factor and Type 1 Collagen Inducible Protein (VCIP) IRES. HEK293T STING Null Cells (293 -Dual™ Null Cells, InvivoGen) were used, which do not contain endogenous STING, and express secreted embryonic alkaline phosphatase (SEAP), placed under the control of the endogenous IFN-stimulated response element (ISRE) promoter, where the coding sequence of ISRE is replaced by the SEAP ORF using knock-in technology. STING activity, thus, is assessed by monitoring type I IFN induced SEAP production. The 293-Dual™ Null cells also express Lucia™ luciferase, a secreted luciferase, placed under the control of the endogenous IFN-P promoter; the coding sequence of IFN- has been replaced by the Lucia™ luciferase ORF using knock-in technology. This allows for the assessment of STING activity by monitoring the expression of IFN-p. Using these cells, STING activity can be assessed by monitoring ISRE-induced SEAP production and/or IFN-P-dependent expression of Lucia™ luciferase. The two reporter proteins, SEAP and Lucia™ Luciferase, can be measured in the cell supernatant using standard assays and detection reagents, such as the QUANTLBlue™ and QUANTI-Luc™ detection reagents (InvivoGen), respectively.

The cells were seeded in 24-well plates coated with poly-L-lysine at 200,000 cells per well, and incubated overnight at 37 °C in a 5% CO2 incubator, to achieve 80% confluency. The following day, 300 ng of each plasmid DNA, and 40 ng of a CMV-GFP vector (i.e., a vector encoding green fluorescent protein under the control of a CMV promoter), were diluted in serum-free media and added to FuGENE® transfection reagent (Promega), at the proper reagent:DNA ratios, with untransfected wells serving as negative controls (in duplicates). Cell culture supernatants from each sample were collected 48 hours post-transfection. The STING activity of the huSTINGN154S/R284G tazCTT variant was evaluated using the ISRE-SEAP and IFN-P-Lucia™ reporter systems. The type I interferon (IFN) activity (induced by STING) was assessed by monitoring type I IFN- stimulated SEAP production in the cell supernatants. 20 pL of cell culture supernatant was added to 180 pL of QUANTI-Blue™ reagent (InvivoGen), which is used for measuring SEAP. Type I interferon activation was determined by measuring ISRE- induced SEAP activity on a SpectraMax® M3 Spectrophotometer (Molecular Devices), at an absorbance wavelength of 650 nm. The type I interferon (IFN) activity (induced by STING) also was assessed by monitoring the type I IFN-stimulated Lucia™ luciferase production in the cell supernatants. 20 pL of cell culture supernatant was added to 50 pL of QUANTI-Luc™ reagent (InvivoGen), which is used for measuring Lucia™ luciferase activity. Type I interferon activation was determined by measuring IFN -induced Lucia™ luciferase activity on a SpectraMax® M3 Spectrophotometer (Molecular Devices), on the luminescence setting.

Cell culture supernatants also were assessed for the expression of human CTLA4 scFv-Fc. Human CTLA4 (R&D Systems) was coated overnight at 4 °C on a high protein-binding 96-well plate, at a concentration of 1 ug/ml. The wells were then washed three times with PBS 0.05% Tween-20, and the wells were blocked with ELISA blocking buffer for 1 hour at room temperature. The wells were then washed three times with PBS 0.05% Tween-20. HEK293T cell culture supernatants, containing anti-CTLA4 scFv-Fc were added to the wells and incubated for 2 hours at room temperature. The wells were washed three times with PBS 0.05% Tween-20, and horseradish peroxidase (HRP)-conjugated anti-human antibody was added for a 1 hour incubation. The wells were then washed three times with PBS 0.05% Tween-20, and detection reagent (3,3',5,5'-tetramethylbenzidine (TMB), Thermo Fisher Scientific) was added to the wells. The enzymatic reaction was stopped with sulfuric acid (BioLegend), and the optical densities were read at 450 nm.

The expression results of the human anti-CTLA4 scFv-Fc + hSTING N154S/R284G tazCTT constructs are shown below. Constructs containing both payloads retained expression of each protein. The human anti-CTLA4 scFv-Fc T2A hSTING N 154S/R284G tazCTT construct had better STING activity without the VCIP IRES. Expression of anti-CTLA4 scFv-Fc was higher by ELISA from the construct without VCIP IRES, demonstrating that individual targets can require unique elements for improved expression.

Human anti-CTLA4 scFv-Fc + STING Expression Normalized by Transfection Efficiency

EXAMPLE 44

Bacterial Strain Infection of Macrophages Allows the Presentation of Immunogenic Peptide by MHC-I Molecules

This Example shows that the bacteria provided herein bacteriofect antigen presenting cells with either mRNA delivery or plasmid DNA delivery, and the resulting expressed product is presented by antigen presenting cells. The anti- SIINFEKL/H-2Kb antibody, used for this demonstration, binds specifically to the ovalbumin SIINFEKL [SEQ ID NO:492] peptide only when presented by MHC-I (H- 2Kb) at the surface of IC21 cells. This demonstrates that the vaccines strains provided herein, such as those that encode STING + IL-15 receptor complex + cancer antigen, can treat immune desert tumors, also can could be used as a cancer vaccine in the neoadjuvant setting, also can be used as a platform to present pathogenic antigens for prophylactic vaccination, and can be used as a treatment for pathogen infection, such as, for example, MRSA, chronic viral hepatitis infection, chronic p.gingivalis, HIV, and other chronic infections, as well as acute infection.

In demonstrating the above, the experiment in this Example, shows how the infection of antigen-presenting cells by the exemplary strain, designated YS 6 6Aasd//^LG/ApagP/AansB/AcsgD , leads to presentation of immunogenic peptides by the major histocompatibility complex class-I (MHC-I) to CD8+ T cells. YS1646Aasd/AFLG/ApagP/AansB/AcsgD strains encoding the chicken ovalbumin under the eukaryotic CMV promoter and the prokaryotic MTL promoter followed or not with the encephalomyocarditis virus (EMCV) internal ribosome entry site (IRES) were generated The chicken ovalbumin is a model antigen, and the immune- dominant SIINFEKL peptide derived from the chicken ovalbumin was monitored for its presentation by the murine H-2Kb MHC-I molecules at the surface of antigen- presenting cells. The murine peritoneal macrophages cell line IC21 (ATCC) was used as antigen-presenting cell in the assay.

Fifty thousand IC21 cells per well of a 96-well flat-bottom plate were seeded in lOOul RPMI 10% FBS and left in a 37°C and 5% CO2 incubator over-night. The next day, IC21 cells were infected with YS1646Aasd/AFLG/ApagP/AansB/AcsgD strains containing plasmids encoding the chicken ovalbumin under the regulation of the CMV or MTL promoters. The bacterial strains were added to the cultured cells and the plate was centrifugated at 500g for 5 minutes at room temperature. Cells were then incubated for Ih at 37°C and culture media containing gentamycin was added for a final concentration of 50ug/ml. Infected cells were left in at 37°C and 5% CO2 incubator for seven hours. Seven hours after the infection, IC21 cells were detached using PBS lOmM EDTA and stained with anti-murine F4/80 APC-conjugated antibodies and anti-SIINFEKL/H-2Kb PE-conjugated antibodies (both from Biolegend), and the data was acquired using the ACEA Novocyte flow-cytometer.

The anti-SIINFEKL/H-2Kb antibody binds specifically to the ovalbumin SIINFEKL peptide only when presented by MHC-I (H-2Kb) at the surface of IC21 cells. The mean fluorescence intensity for the H-2Kb/SIINFEKL presentation by IC21 cells seven hours post-infection was measured. The results are presented in the table below, which shows the net mean fluorescence intensity (Net MFI) observed in the groups treated with the ovalbumin encoding strains over the uninfected cells.

*SEM = standard error of the mean

This Example shows the ability of strains, such as 646l asd/ ¥ G/ApagP/AansB/AcsgD, to deliver antigens that are processed by the antigen-presenting cells and presented to CD8+ T cells upon delivery, to mount a T cell specific response against antigens relevant to or from pathogens or cancer cells.

EXAMPLE 45

Infection of Immunostimulatory Bacteria Transformed with Plasmids Encoding huIL-15Ra-IL-15sc and huSTING N154S/R284G tazCTT Converts Human M2 Macrophages, and Not Ml Macrophages, into a Hybrid Ml/M2-Like Phenotype That Retains M2 Phagocytic Functions

A long-held paradigm of macrophage phenotypes in tissues describes two distinct phenotypes: Ml and M2. An Ml macrophage phenotype is activated by sensing of foreign pathogens and/or damaged tumor cells to produce pro- inflammatory cytokines, such as IFNy, TNFoc, IL-6, IL-ip, IL-12 and IL-23, and chemokines such as CXCL10 and CXCL11, that allegedly recruit and activate adaptive immune cells to promote adaptive immunity. Direct killing by Ml macrophages can occur through production of IFNy and cellular damage through production of reactive-oxygen species (ROS; Anfray etal., (2019) Cells 9:46). Chronic inflammation is understood to promote immunosuppressive M2 macrophage differentiation to resolve the tissue damage and promote wound healing. M2 macrophages produce immunosuppressive cytokines such as IL-10 and TGF0, and tissue remodeling factors such as vascular endothelial growth factor (VEGF) and matrix metalloproteases (MMPs), in order to promote angiogenesis and tissue repair. Because of these immunosuppressive functions of M2, including purported upregulation of the immune checkpoints PD-1 and PD-L1 that impair T-cell function, they are primarily responsible for suppressing adaptive immunity in the tumor microenvironment (see, e.g., Anfray et al., (2019) Cells 9:46). According to this paradigm, it is Ml macrophages that induce tumor cell apoptosis via FasL, and phagocytosis of tumor cells (efferocytosis) and pathogens, as well as antigen presentation to T-cells. As such, many immunotherapies focus on repolarizing tumor M2 macrophages to Ml macrophages.

There are problems with this paradigm. First, if Ml cells are phagocytic and not M2, it is unclear how M2 macrophages resolve tissue damage if they cannot phagocytose dying cells. Similarly, there is a question regarding how Ml macrophages promote T-cell activation when they do not produce type I IFN, and instead produce pro-inflammatory cytokines that recruit and activate innate cells such as neutrophils and myeloid-derived suppressor cells that are unable to present antigens to T-cells. These discrepancies are highlighted by hospitalized SARS-CoV-2 patients, where lung immunophenotyping and serum cytokine analyses of patients with advanced lung disease have concluded that the production of pro-inflammatory cytokines from infiltrating Ml macrophages result in lung tissue destruction, impaired type I IFN production and CD8 ⁺ T-cell priming, and suppression of viral clearance, leading to disease progression and lethal lung damage (Gracia-Hernandez etal., (2020) Front Pharmacol 77:577571).

To further elucidate the role of Ml and M2 macrophages in anti -turn or immunity, and the effects on macrophage polarization from infection with immunostimulatory bacteria provided herein, such the immunostimulatory bacteria that carry a plasmid encoding huIL-15Ra-IL-15sc and huSTINGN154S/R284G tazCTT, experiments were conducted in which primary human monocyte-derived macrophages, differentiated into Ml and M2 macrophages, were assessed postinfection for Ml and M2 markers; results are summarized in the table below.

Table: Markers Attributed to Ml and M2 Macrophage Phenotypes

A study, described above, and in co-owned International PCT publication No.

WO2021/097144), infection with the immunostimulatory bacterium designated YS1646Aas<7/ AFI.G/ \pagPi \ansB \csgl) strain and containing a plasmid encoding one or more immunostimulatory proteins, converts an M2 phenotype macrophage into an Ml or Ml -like macrophage, capable of upregulating co-stimulatory molecules, the CCR7 marker of lymph node migration, T-cell recruiting chemokines CXCL10 and CXCL11, as well as production of type I IFN. Since the prior study characterized polarization from an M2 macrophage to an Ml phenotype, a further study was performed to observe the effect of the YS1646Aasc7/AFLG/A/rag7’/AcOTsB/AcsgD strain containing a plasmid encoding huIL-15Ra-IL-15sc + huSTING N154S/R284G tazCTT on Ml and M2 phenotypes following infection.

Frozen human monocytes, isolated from healthy human donors, were thawed in complete medium (RPMI-1640 + 10% FBS), and washed by washed by centrifugation for 10 minutes at 600xg at room temperature. Monocytes were resuspended in ImmunoCult™- SF Macrophage Medium (StemCell Technologies) containing 100 ng/mL human M-CSF. Monocytes (2.5e5 per well) then were seeded in a 48-well plate with a final volume of 500 pL. Five days later (on day 5), fresh medium with cytokines was added to generate Ml and M2 macrophages. To generate primary human Ml macrophages, 250 pL of ImmunoCult™- SF Macrophage Medium containing 10 ng/mL LPS + 50 ng/mL IFNy was added per well and incubated for 48 hours. To generate primary human M2 macrophages, 250 pL of ImmunoCult™- SF Macrophage Medium containing 150 ng/mL human M-CSF + 60 ng/mL human IL-4 + 60 ng/mL human IL- 10 was added per well and incubated for 48 hours.

On day 7, duplicate wells were infected at an MOI of 10, for two hours in RPMI, with the YS1646Aasz//AFLG/A/iag /Aans /Acs Z) strain containing a plasmid encoding huIL-15Ra-IL-15sc + huSTING N154S/R284G tazCTT. The cells then were washed 3 times with PBS and resuspended in RPMI + 100 pg/mL gentamicin (Sigma). After 48 hours, the cells either were collected for gene expression analysis using qPCR analysis (as previously mentioned) or for cell surface protein expression analysis using flow cytometry.

For qPCR analysis, the cells were collected for RNA isolation using the RNeasy® Plus Mini Kit (Qiagen), according to the manufacturer’s instructions. Total RNA concentration was measured using a NanoDrop™ 2000 UV-Vis Spectrophotometer (Thermo Scientific). The purity of each sample also was assessed from the A260/A230 absorption ratio. Synthesis of cDNA was performed from 0.5-1 pg of template RNA using a CFX96™ Real-Time System (Bio-Rad) and i Script™ Reverse Transcription Supermix for RT-qPCR (Bio-Rad) in a 20 pL reaction, according to the manufacturer’s instructions. qPCR was performed with a CFX96 Real-Time System (Bio-Rad). Custom qPCR primers for IL-15 (For: 5’- GCAAGAGGAATTTGCCATCG; Rev: 5’-TTGCTCCATCCCGCTAATG) and STING (For: AACTCTGGTTTCAAGCGTAAAG; Rev: CAGGGCAGGGTCTCTAATG), SEQ ID NOS: 496-499, respectively, were prepared.

The qPCR reaction (20 uL) was conducted per protocol, using either the SsoAdvanced Universal SYBR® Green Supermix or the iQ Multiplex Powermix (Bio-Rad). The standard thermocycling program on the Bio-Rad CFX96 Real-Time System included a 95 °C denaturation for 150 sec, followed by 39 cycles of 95 °C for 15 sec and 60 °C for 55 sec. Quantification of the target mRNA was normalized using Actin reference mRNA (Bio-Rad, qHsaCEP0036280). AC _q was calculated as the difference between the target and reference gene.

As shown in the table below, expression of the IL- 15 (IL-15/IL-15R alpha chain complex) and STING was compared among untreated Ml and M2 macrophages controls, the YS1646 Aa«Z/ AFLG/ A/ragP/AawxS/AcsgD strain containing a plasmid control, and YS I 646Aa.S6// AFLG/ kpagPI \ansB \csgl) strain containing a plasmid encoding huIL-15Ra-IL-15sc + huSTINGN154S/R284G tazCTT expression on macrophages. STACT in this table, and those below, refers to the immunostimulatory bacterium strain YS1646Aasz// AFL

The data show that the Ml macrophages did not express the plasmid-encoded payloads. Since, it was found that Ml macrophages were not capable of expressing plasmid payloads following infection, the next study focused on characterizing the M2-specific phenotypes of M2 macrophages following infection with the bacteria. For this, a study was performed to determine the phenotype of primary human M2 macrophages infected with the YS1646 \cisdl AFLG/ kpagPI kansB/kcsgD strain containing a plasmid encoding huIL-15Ra-IL-15sc and huSTING N154S/R284G tazCTT. As shown below, the M2 macrophage were converted to a hybrid M1/M2 phenotype.

Frozen human monocytes, isolated from healthy human donors, were thawed in complete medium (RPMI-1640 + 10% FBS), and washed by washed by centrifugation for 10 minutes at 600xg at room temperature. Monocytes were resuspended in ImmunoCult™- SF Macrophage Medium (StemCell Technologies) containing 100 ng/mL human M-CSF. Monocytes (2.5e5 per well) then were seeded in a 48-well plate with a final volume of 500 pL. Five days later (on day 5), fresh medium with cytokines were added to generate M2 macrophages. To generate primary human M2 macrophages, 250 pL of ImmunoCult™- SF Macrophage Medium containing 150 ng/mL human M-CSF + 60 ng/mL human IL-4 + 60 ng/mL human IL- 10 was added per well and incubated for 48 hours.

On day 7, duplicate wells were infected at an MOI of 10, for two hours in RPMI, with the YS1646AazZ/AFLG/A/ /Aans /Acs £> strains containing plasmids encoding human 0-actin as a plasmid control, huIL-15Ra-IL-15sc alone, huSTING N154S/R284GtazCTT alone, or huIL-15Ra-IL-15sc + huSTING. The cells then were washed 3 times with PBS and resuspended in RPMI + 100 pg/mL gentamicin (Sigma). As controls, human Ml macrophages were generated in 10 ng/mL LPS + 50 ng/mL IFNy, and human M2 macrophages were treated either alone or in combination with recombinant huIL-15 or the small molecule STING agonist Rp,Rp, a synthetic analog of c-di-AMP with a 3’5’ mixed linkage and a disulfide bridge in the R,R confirmation (Invivogen).

After 48 hours, supernatants were harvested and evaluated for downstream signaling differences using a custom human M2/M1 U-Plex® panel (Meso Scale Discovery), according to the manufacturer’s protocol. Cytokine secretion was measured, and the average of the duplicate was calculated. The cells were either collected for gene expression analysis using qPCR analysis or cell surface protein expression analysis using flow cytometry.

For qPCR analysis, the cells were collected for RNA isolation using the RNeasy® Plus Mini Kit (Qiagen), according to the manufacturer’s instructions. Total RNA concentration was measured using a NanoDrop™ 2000 UV-Vis Spectrophotometer (Thermo Scientific). The purity of each sample also was assessed from the A260/A230 absorption ratio. Synthesis of cDNA was performed from 0.5-1 pg of template RNA using a CFX96™ Real-Time System (Bio-Rad) and i Script™ Reverse Transcription Supermix for RT-qPCR (Bio-Rad) in a 20 pL reaction, according to the manufacturer’s instructions. qPCR was performed with a CFX96 Real-Time System (Bio-Rad). PrimePCR™ Probe Assay for huCD80 (qHsaCIP0026764), huCD86 (qHsaCEP0050792), huCCR7 (qHsaCIP0033364), huCXCLIO (qHsaCEP0053880), huCXCLl l (qHsaCEP0024091), huCD14 (qHsaCEP0053971), huCD206 (qHsaCIP0039797), CD209 (qHsaCEP0049877), huCD68 (qHsaCED0007025) and huCD163 (qHsaCIP0028398) (All from Bio-Rad). The qPCR reaction (20 pL) was conducted per protocol, using either the SsoAdvanced Universal SYBR® Green Supermix or the iQ Multiplex Powermix (Bio-Rad). The standard thermocycling program on the Bio-Rad CFX96 Real-Time System consisted of a 95 °C denaturation for 150 sec, followed by 39 cycles of 95 °C for 15 sec and 60 °C for 55 sec. Quantification of the target mRNA was normalized using Actin reference mRNA (Bio-Rad, qHsaCEP0036280). AC _q was calculated as the difference between the target and reference gene.

For flow cytometry analysis, cells were washed with 500-uL of DPBS and detached by incubating with 10-mM EDTA in PBS on ice for 20-min. Lifted cells were washed once with flow buffer (PBS + 2% FBS) and resuspended in 100-uL of LIVE/DEAD Violet Fixable Viability Dye. After 20-min incubation at RT, cells were washed once with flow buffer and resuspended in 50 pL of flow buffer containing True-Stain Monocyte Blocker and Human TruStain FcX. After 10-min incubation on ice, cell surface flow cytometry antibodies CD80 BV605 clone 2D10; CD14 BV650 clone M5E2; PD-L1 BV785 clone 2E9.2A3; CCR7 AF488 cloneG043H7; CD206 PE-CF594 clone 15-2; PD-1 PE-Cy7 clone EH12.2H7; CD209 APC clone 9E9A8; and CD 163 APC-Cy7 clone GHI/61 (all from BioLegend) were added. After 20-min of incubation on ice and in the dark, cells were washed twice with flow buffer. Cells then were resuspended in flow buffer and data were acquired using the NovoCyte® flow cytometer (ACEA Biosciences, Inc.) and analyzed using FlowJo™ software (Tree Star, Inc.).

As shown in the table below, compared to the untreated M2 macrophages control, all YS l646A .sc// AFLG/ pagPI ansB/ csgD strains containing plasmid- encoded payloads induced high expression of the Ml -like macrophage markers CD80, CD86, CCR7, HLA-A, CXCL10 and CXCL11. These macrophages also induced expression of IFNP, with the highest expression in the huIL-15Ra-IL-15sc + huSTING N 154S/R284G tazCTT group, which was higher than either the huSTING N 154S/R284G tazCTT strain alone, or M2 macrophages treated with the small molecule STING agonist RpRp, demonstrating payload-dependent effects. Importantly, all bacterially infected M2 macrophages maintained high levels of M2-like surface expression of pattern recognition receptors and scavenger receptors, such as CD14, CD163 and CD209, as compared to Ml macrophages. Importantly, expression of the costimulatory receptor CD80 was highest in the huSTING N 154S/R284G tazCTT and huIL-15Ra-IL-15sc + huSTING N154S/R284G tazCTT strains, and the huIL-15Ra- IL-15sc + huSTING N154S/R284G tazCTT strain had the highest IFNp and CD86 co-

5 stimulatory expression, also demonstrating payload dependence for these targets.

* ST ACT refers to the immunostimulatory bacteria

The table below outlines the changes in phenotype of these hybrid macrophages. Interestingly, markers of phagocytosis such as CD209 were initially higher in M2 macrophages and not Ml, and infection retained expression of this 0 marker.

Table: Markers Attributed to Ml and M2 Macrophage Phenotypes Post- Treatment

The next experiment assessed whether Ml macrophages lacking CD209 expression, or M2 macrophages with CD209 expression, are capable of phagocytosing apoptotic tumor cells, and the impact infection has on M2 macrophage phagocytic and efferocytic functions. For this, Ml and M2 macrophages were differentiated, and M2 macrophages infected with the YS I646Aa.stZ ^/ AFLG/A/?a /7Aa//.sA Ac.sg/.) strains containing plasmid payloads were assessed in phagocytosis of apoptotic tumor cells (efferocytosis) and bacterial phagocytosis assays. Frozen human monocytes, isolated from healthy human donors, were thawed in complete medium (RPMI-1640 + 10% FBS), and washed by washed by centrifugation for 10 minutes at 600xg at room temperature. Monocytes were resuspended in ImmunoCult™- SF Macrophage Medium (StemCell Technologies) containing 100 ng/mL human M-CSF. Monocytes (2.5e5 per well) then were seeded in a 48-well plate with a final volume of 500 pL. Five days later (on day 5), fresh medium with cytokines were added to generate M2 macrophages. To generate primary human M2 macrophages, 250 pL of ImmunoCult™- SF Macrophage Medium containing 150 ng/mL human M-CSF + 60 ng/mL human IL-4 + 60 ng/mL human IL- 10 was added per well and incubated for 48 hours.

On day 7, duplicate wells were infected at an MOI of 10, for two hours in RPMI, with the following YS I646Aa.sc// AFLG/ pagPI cmsB/ csgD strains containing plasmids encoding human p-actin as a plasmid control, huIL-15Ra-IL- 15sc alone, huSTING N154S/R284G tazCTT alone, or huIL-15Rot-IL-15sc + huSTING. The cells then were washed 3 times with PBS and resuspended in RPMI + 100 pg/mL gentamicin (Sigma). As controls, human Ml macrophages were generated in 10 ng/mL LPS + 50 ng/mL IFNy, and human M2 macrophages were treated either alone or in combination with recombinant huIL-15 or the small molecule STING agonist Rp,Rp.

After 48 hours, 1.5e5 CFSE-labeled apoptotic Raji (induced by 24 hours 1- mM H2O2 treatment) or pHrodo™ Green E. coli BioParticles™ Conjugate (Invitrogen) were added to each well and incubate at 37 °C. After 3 hours, the cells were washed with 500-uL of DPBS and detached by incubating with 10-mM EDTA in PBS on ice for 20-min. Lifted cells were washed once with flow buffer (PBS + 2% FBS) and resuspended in 50 pL of flow buffer containing True-Stain Monocyte Blocker and Human TruStain FcX. After 10-min incubation on ice, cell surface flow cytometry antibody SIRPa PE-Cy7 clone SE5A5 (BioLegend) were added. After 20- min of incubation on ice and in the dark, cells were washed twice with flow buffer. Cells then were resuspended in flow buffer and data were acquired using the NovoCyte® flow cytometer (ACEA Biosciences, Inc.) and analyzed using FlowJo™ software (Tree Star, Inc.). SIRPa-positive population was gated to differentiate macrophages from apoptotic Raji and pHrodo™ Green E. coli BioParticles™ Conjugate.

As shown in the table below, compared to untreated M2 macrophages control, all YS1646A«5t//AFLG/A/?tzg7 ^? /Aan5B/Ac5gZ) strains infected M2 macrophages had decreased apoptotic Raji phagocytosis capacity, but maintained high phagocytosis capacity when compared to the untreated Ml macrophages control. In addition, compared to other strains, the YS I 646At/W/ AFLG/ \pagPI /^ansB/l\csgD strain containing a plasmid encoding huIL-15Ra-IL-15sc + huSTINGN154S/R284G tazCTT induced the highest bacteria bioparticle conjugate phagocytosis capacity.

These data demonstrate the ability of immunostimulatory bacterial strains provided herein, such as YSl646Basd/A¥LG/dpagP/dansB/dcsgD’ containing a plasmid encoding huIL-15Ra-IL-15sc + huSTING N154S/R284G tazCTT or similar payload, to transform M2 macrophages into a hybrid M1/M2 macrophage phenotype that maintain M2-like scavenging receptors and phagocytic ability.

EXAMPLE 46

Macrophage Phagocytosis and Cell Cycle Progression Are Required for Payload Delivery Following Infection of TLR 2/4/5 Attenuated Immunostimulatory Bacterial Strains Containing Plasmid Payloads

The finding in the Example above that Ml macrophages do not express the encoded payloads and that only M2 macrophages are able to express payloads following infection, prompted further questions and review to understand how and why the differentiation conditions result in impaired payload delivery or expression in Ml macrophage. Proliferating macrophages have been described in fibrotic diseases such as idiopathic pulmonary fibrosis (IPF) (Morse et al., (2019) Eur Respir J 54(2): 1802441). Factors associated with M2 macrophage polarization, such as M- CSF, phagocytosis and efferocytosis, promote macrophage self-renewal (proliferation). Treatment with pro-inflammatory cytokines and other bacterial components associated with Ml polarization, such as JFNy and lipopolysaccharide (LPS) from Gram-negative bacteria, as well as necrotic, hypoxic and adenosine-rich environments, inhibit cell cycle prevention (Roszer et al., (2018) Cells 7(8): 103). The lack of proliferation of Ml macrophages makes sense from functional and metabolic standpoint, where the need for enhanced wound healing and phagocytic functions requires more macrophages, whereas bacterial pathogen sensing and production of pro-inflammatory cytokines is metabolically taxing and prevents cell cycle machinery, and also from the fact that Ml macrophages are terminally differentiated, and thus are not proliferating.

Proliferation is a requirement for plasmid gene transfer. The ability of transgene expression following transfection of plasmid DNA, where the DNA must enter the nucleus and be transcribed into mRNA, requires spindle breakdown to provide nuclear entry (Mortimer et al., (1999) Gene Therapy 6:403-411). Gene delivery into cells to facilitate plasmid transfer and gene expression is inefficient in non-dividing cells. Those in the gene therapy field pursued virally-delivered nucleic acids that integrate into the nucleus without requiring cell division (Kirchenbuechler et al, (2016) Exp. Cell Res. 345(1): 1-15).

To assess the impact of bacterial sensing on Ml polarization, cell cycle inhibition, and the ability to express plasmid payloads, infections were performed in primary M2 macrophages with a series of the TLR-attenuated strains of the immunostimulatory bacteria provided herein. These were assessed for their ability to provide plasmid transfer following infection and expression of the secreted Nanoluciferase® (NanoLuc®, Promega) protein. For this, the parental YS1646 strain, which is a TLR2/4/5-intact strain, was compared to the TLR5 attenuated strain YS IO Ac/.stZAFLG, the TLR4 and TLR5 attenuated strain YS I646 €/.s< ^/ AFLG/A/?<:/ / ^> , and the TLR2/4/5 attenuated strain

YS I646Atf.st/AFLG/A/2tf _< p/7A<:7//.sA Ac.s /J, all containing the NanoLuc® plasmid.

Frozen human monocytes, isolated from healthy human donors, were thawed in complete medium (RPMI-1640 + 10% FBS), and washed by washed by centrifugation for 10 minutes at 600xg at room temperature. Monocytes were resuspended in ImmunoCult™- SF Macrophage Medium (StemCell Technologies) containing 100 ng/mL human M-CSF. Monocytes (2.5e5 per well) then were seeded in a 48-well plate with a final volume of 500 pL. Five days later (on day 5), fresh medium with cytokines were added to generate M2 macrophages. To generate primary human M2 macrophages, 250 pL of ImmunoCult™- SF Macrophage Medium containing 150 ng/mL human M-CSF + 60 ng/mL human IL-4 + 60 ng/mL human IL- 10 was added per well and incubated for 48 hours.

On day 7, duplicate wells were infected at an MOI of 10, for two hours in RPMI, with the YS1646 and YS1646 TLR-attenuated strains containing a plasmid encoding NanoLuc®. The cells then were washed 3 times with PBS and resuspended in RPMI + 100 pg/mL gentamicin (Sigma). After 48 hours, the cells were collected for gene expression analysis using qPCR analysis (described above).

For qPCR analysis, the cells were collected for RNA isolation using the RNeasy® Plus Mini Kit (Qiagen), according to the manufacturer’s instructions. Total RNA concentration was measured using a NanoDrop™ 2000 UV-Vis Spectrophotometer (Thermo Scientific). The purity of each sample also was assessed from the A260/A230 absorption ratio. Synthesis of cDNA was performed from 0.5-1 pg of template RNA using a CFX96™ Real-Time System (Bio-Rad) and i Script™ Reverse Transcription Supermix for RT-qPCR (Bio-Rad) in a 20 pL reaction, according to the manufacturer’s instructions and the average of the triplicate was calculated. qPCR was performed with a CFX96 Real-Time System (Bio-Rad). Custom primers were used for NanoLuc® detection. The qPCR reaction (20 pL) was conducted per protocol, using the SsoAdvanced Universal SYBR® Green Supermix (Bio-Rad). The standard thermocycling program on the Bio-Rad CFX96 Real-Time System consisted of a 95 °C denaturation for 150 sec, followed by 39 cycles of 95 °C for 15 sec and 60 °C for 55 sec. Quantification of the target mRNA was normalized using Actin reference mRNA (Bio-Rad, qHsaCEP0036280). AAC _q was calculated as the difference between the treatments and the uninfected control. As shown in the table below, compared to the untreated M2 macrophages or the parental YS1646 TLR2/4/5 -intact strain, the TLR5 attenuated strain YS1646Aasz//AFLG and the TLR4 and TLR5 attenuated strain YS I646Ac/.s<7AFLG/A/?c/g/ ^> all containing the NanoLuc® plasmid, induced NanoLuc® expression. The TLR2/4/5 attenuated strain YS1646Aas6?/AFLG/A/?ag7’/Aans _J 8/Acsg£>, containing the NanoLuc® plasmid induced significantly higher expression of NanoLuc, by more than 6-fold.

In a separate experiment, human THP1 monocytes were induced to become more phagocytic and M2 -like by the addition of M2 differentiation medium with the addition of osteopontin (SPP1) to enhance wound-healing macrophage properties, in order to assess whether these culture conditions could enhance phagocytosis and proliferation and therefore payload delivery. For this, human monocytic THP-1 cells were differentiated by PMA treatment for 3 days, followed by 2 days of culturing in either media-only or M2-differentiation media. For PMA treatment, THP-1 cells were cultured in RPMI-1640+ 10% FBS medium + 50-ng/mL of PMA for 3 days. On day 3, media was aspirated and replace with either media or media + 20 ng/mL IL-4 + 20 ng/mL IL-10 + 500 ng/mL human osteopontin. On day 5, cells were infected in duplicate at an MOI of 100 using thawed injection stocks, followed by centrifugation at 500 relative centrifugal force (ref) for 5 minutes and incubation at 37 °C for 2 hours. After 2 hours, the cells were washed three times in DPBS and fresh RPML 1640+ 10% FBS medium was added with a gentamycin concentration of 50 pg/ml.

For the phagocytosis assay, 1.5e5 CFSE-labeled apoptotic Raji (induced by 24 hours 1-mM H2O2 treatment) or pHrodo™ Green E. coli BioParticles™ Conjugate (Invitrogen) were added to each well and incubate at 37 °C. After 3 hours, the cells were washed with 500-uL of DPBS and detached by incubating with 10-mM EDTA in PBS on ice for 20-min. Lifted cells were washed once with flow buffer (PBS + 2% FBS) and resuspended in 50 pL of flow buffer containing True-Stain Monocyte Blocker and Human TruStain FcX. After 10-min incubation on ice, cell surface flow cytometry antibody SIRPa PE-Cy7 clone SE5A5 (BioLegend) were added. After 20- min of incubation on ice and in the dark, cells were washed twice with flow buffer. Cells then were resuspended in flow buffer and data were acquired using the NovoCyte® flow cytometer (ACEA Biosciences, Inc.) and analyzed using FlowJo™ software (Tree Star, Inc.). SIRPa-positive population was gated to differentiate macrophages from apoptotic Raji and pHrodo™ Green E. coli BioParticles™ Conjugate. Supernatant osteopontin concentrations were measured using the Human Osteopontin Quantikine® ELISA Kit (R&D Systems).

For qPCR analysis, the cells were collected for RNA isolation using the RNeasy® Plus Mini Kit (Qiagen), according to the manufacturer’s instructions. Total RNA concentration was measured using a NanoDrop™ 2000 UV-Vis Spectrophotometer (Thermo Scientific). The purity of each sample also was assessed from the A260/A230 absorption ratio. Synthesis of cDNA was performed from 0.5-1 pg of template RNA using a CFX96™ Real-Time System (Bio-Rad) and iScript™ Reverse Transcription Supermix for RT-qPCR (Bio-Rad) in a 20 pL reaction, according to the manufacturer’s instructions, and the average of the triplicate was calculated. qPCR was performed with a CFX96 Real-Time System (Bio-Rad). Custom primers were used forNanoLuc® detection. The qPCR reaction (20 pL) was conducted per protocol, using the SsoAdvanced Universal SYBR® Green Supermix (Bio-Rad). The standard thermocycling program on the Bio-Rad CFX96 Real-Time System consisted of a 95 °C denaturation for 150 sec, followed by 39 cycles of 95 °C for 15 sec and 60 °C for 55 sec. Quantification of the target mRNA was normalized using Actin reference mRNA (Bio-Rad, qHsaCEP0036280). AAC _q was calculated as the difference between the treatments and the uninfected control.

As shown in the tables below, compared to PMA-treated THP-1 cells, the addition of M2-differentiation medium containing osteopontin increased THP-1 phagocytosis capacity as well as its osteopontin secretion. Gene expression of the NanoLuc® payload, however, was not detected, demonstrating an inability to express payloads, despite having an M2 -like macrophage phenotype.

The finding that culture conditions designed to promote a M2-like macrophage with enhanced phagocytic functions still prevents plasmid expression was not expected, and prompted further queries, which revealed that the PMA treatment was responsible. In a published experiment, PMA-treated THP1 cells were assessed over a 96-hour time course by Cap Analysis of Gene Expression (CAGE) for transcriptional regulation. The PMA-treated cells became adherent as expected, but the numbers of cells entering S phase and exiting G2/M was significantly reduced. Within 2 hours, early macrophage differentiation genes such as MAFB were induced, and cell proliferation genes such as MYC were reduced (Gazova el al., (2020) Front. Cell Dev. Biol. 8:498). The transcription factors Maf-B and c-Maf have been described to control macrophage self-renewal, with upregulation impairing proliferation, dependent on suppression of MYC and other proliferation markers such as the mitotic checkpoint gene BUB 1 (Aziz etal., (2009) Science 326:867:71; Soucie e/ aZ, (2016) Science 351(6274):aad5510). In human macrophages treated with agonists of the cytosolic RNA sensors RIG-I and MDA5, expression of MAFB suppressed the production of type I IFN (Kim etal., (2010) Nat Immunol . 11(8):743-50).

In order to determine whether PMA treatment was similarly impairing THP1 proliferation, untreated THP1 cells were compared to PMA-treated cells and assessed for gene expression of proliferation markers. For this, THP-1 cells were cultured in either the presence or absent of 50 ng/mL PMA for 48 hrs. Cells were collected for RNA isolation using the RNeasy® Plus Mini Kit (Qiagen), according to the manufacturer’s instructions. Total RNA concentration was measured using a NanoDrop™ 2000 UV-Vis Spectrophotometer (Thermo Scientific). The purity of each sample also was assessed from the A260/A230 absorption ratio. Synthesis of cDNA was performed from 0.5-1 pg of template RNA using a CFX96™ Real-Time System (Bio-Rad) and i Script™ Reverse Transcription Supermix for RT-qPCR (BioRad) in a 20 pL reaction, according to the manufacturer’s instructions. qPCR was performed with a CFX96 Real-Time System (Bio-Rad). PrimePCR™ Probe Assay for huMAFB (qHsaCEP0024306), huMAF (qHsaCEP0054577), huSTMNl (qHsaCIP0041274), huMYC (qHsaCIP0028650) and huBUBl (qHsaCIP0033109) (All from Bio-Rad). The qPCR reaction (20 pL) was conducted per protocol, using either the SsoAdvanced Universal SYBR® Green Supermix or the iQ Multiplex Powermix (Bio-Rad). The standard thermocycling program on the Bio-Rad CFX96 Real-Time System consisted of a 95 °C denaturation for 150 sec, followed by 39 cycles of 95 °C for 15 sec and 60 °C for 55 sec. Quantification of the target mRNA was normalized using Actin reference mRNA (Bio-Rad, qHsaCEP0036280). AC _q was calculated as the difference between the target and reference gene.

As shown in the table below, PMA treatment suppresses proliferation- associated genes such as MYC and BUB1. The proliferation-inhibiting transcription factors MAFB and MAF were induced by PMA treatment, blocking cell cycle progression. These data further show the importance of promoting cell cycle in order to transfer plasmids and express encoded payloads.

A follow-up study was performed to determine whether THP1 cells cultured with M-CSF could promote proliferation and payload delivery, as compared to the PMA-treated terminally differentiated THP-1 cells. Both were infected with the YS1646 asd/ AFLG/ ApagP/AansB/ csgD strain containing a plasmid encoding huIL- 15Ra-IL-15sc + huSTINGN154S/R284GtazCTT, or the NanoLuc® plasmid control, and compared to uninfected but differentiated THP1 cells for evidence of payload delivery

THP-1 cells were either terminally differentiated by PMA treatment prior to infection with the immunostimulatory bacteria, or treated with M-CSF to promote cell cycle progression after infection. For PMA treatment, THP-1 cells were cultured in RPMI-1640+ 10% FBS medium + 50 ng/mL of PMA for 5 days. On day 5, duplicate wells were infected at an MOI of 100, for one hour in RPMI, with the YS1646Aa«// AFLG/ ApagP/ AansB Ac gl) strain containing a plasmid encoding huIL- 15Ra-IL-15sc + huSTINGN154S/R284GtazCTT. The cells then were washed 3 times with PBS and resuspended in RPMI + 100 pg/mL gentamicin (Sigma). For the M-CSF-treated THP-1 cells still in suspension, THP-1 cells in 5-mL tubes were infected at an MOI of 100, for one hour in RPMI, with the YS1646Aasz// AFLG/ ApagP/ AansB/AcsgD strain containing a plasmid encoding huIL- 15Ra-IL-15sc + huSTINGN154S/R284GtazCTT. The cells were then supplemented with RPMI- 1640+ 10% FBS medium at a final concentration of 100 pg/mL gentamicin (Sigma) and 100 ng/ml M-CSF. After 48 hours, supernatants were harvested, cleared of cell debris, and analyzed for cytokine profile using a U-Plex® (Meso Scale Discovery).

As shown in the table below, infected THP1 monocytes that were treated with M-CSF, but not PMA, responded to infection in a payload-dependent manner. IFN-P, CXCLlO/IP-10, and SPP1/OPN were highest after infection with YS1646Aa«// AFLG/ pagPI ansB// csgD strain containing a plasmid encoding huIL- 15Ra-IL-15sc + huSTINGN154S/R284GtazCTT, as compared to the Nanoluc® control or the uninfected, but only if the cells were pretreated with M-CSF, as PMA ablated the payload response. IL-6 displayed a payload-dependent response only with the M-CSF-treated cells. These data demonstrate that cells that are differentiated to promote proliferation express plasmid-encoded payloads.

In a previous biodistribution study performed with an immunostimulatory bacteria provided herein (see, co-owned International PCT Publication No. WO2021/097144) containing the plasmid payload expressing NanoLuc®, it was demonstrated that while the bacteria could be found at a steady state in the spleens and livers of tumor-colonized mice, the NanoLuc® payload was detected in the tumor. Data and results shown herein show why splenic macrophages and liver Kupffer cells, each of which is highly phagocytic, do not express plasmid-encoded payloads. A study was performed to determine whether primary human M2 macrophages, but not Kupffer or HUVEC, infected with the

YS IbdbAc/.s't// AFLG/ pagPI ansB/ csgD strain containing a plasmid encoding huIL- 15Ra-IL-15sc + huSTINGN154S/R284GtazCTT, induces efficient payload delivery.

Frozen human monocytes, isolated from healthy human donors, were thawed in complete medium (RPML1640 + 10% FBS), and washed by washed by centrifugation for 10 minutes at 600xg at room temperature. Monocytes were resuspended in ImmunoCult™- SF Macrophage Medium (StemCell Technologies) containing 100 ng/mL human M-CSF. Monocytes (2.5e5 per well) then were seeded in a 48-well plate with a final volume of 500 LIL Five days later (on day 5), fresh medium with cytokines were added to generate M2 macrophages. To generate primary human M2 macrophages, 250 pL of ImmunoCult™- SF Macrophage Medium containing 150 ng/mL human M-CSF + 60 ng/mL human IL-4 + 60 ng/mL human IL- 10 was added per well and incubated for 48 hours. Kupffer and HUVEC cells were seeded one day prior to infection, with the Kupffer cells seeded in Collagen-I treated wells.

On day 7, duplicate wells were infected at an MOI of 100, for two hours in RPMI, with the YS1646A '5t//AFLG/A/?tzgZ’/Aan52?/Ac5gD strain containing a plasmid encoding huIL-15Rci-IL-15sc + huSTINGN154S/R284G tazCTT or NanoLuciferase® and compared to the strains encoding huIL-15Ra-IL-15sc or huSTING N154S/R284G tazCTT alone. The cells then were washed 3 times with PBS and resuspended in RPMI + 100 pg/mL gentamicin (Sigma). After 48 hours, supernatants were harvested, cleared of cell debris and analyzed for cytokine profile using a U-Plex® (Meso Scale Discovery).

For qPCR analysis, the cells were collected for RNA isolation using the RNeasy® Plus Mini Kit (Qiagen), according to the manufacturer’s instructions. Total RNA concentration was measured using a NanoDrop™ 2000 UV-Vis Spectrophotometer (Thermo Scientific). The purity of each sample also was assessed from the A260/A230 absorption ratio. Synthesis of cDNA was performed from 0.5-1 pg of template RNA using a CFX96™ Real-Time System (Bio-Rad) and i Script™ Reverse Transcription Supermix for RT-qPCR (Bio-Rad) in a 20 pL reaction, according to the manufacturer’s instructions. qPCR was performed with a CFX96 Real-Time System (Bio-Rad). Custom qPCR primers, described in the Example above, for IL-15 (For: 5’-GCAAGAGGAATTTGCCATCG; Rev: 5’- TTGCTCCATCCCGCTAATG) and STING (For: AACTCTGGTTTCAAGCGTAAAG; Rev: CAGGGCAGGGTCTCTAATG) (SEQ ID NOS: 496-499, respectively) were prepared. The qPCR reaction (20 pL) was conducted per protocol, using either the SsoAdvanced Universal SYBR® Green Supermix or the iQ Multiplex Powermix (Bio-Rad). The standard thermocycling program on the Bio-Rad CFX96 Real-Time System was a 95 °C denaturation for 150 sec, followed by 39 cycles of 95 °C for 15 sec and 60 °C for 55 sec. Quantification of the target mRNA was normalized using Actin reference mRNA (Bio-Rad, qHsaCEP0036280). AC _q was calculated as the difference between the target and reference gene.

As shown in the table below, the immunostimulatory bacteria encoding huIL- 15Ra-IL-15sc + huSTINGN154S/R284GtazCTT induced high level of payload expression in M2 macrophages, but not in Kupffer cells or HUVECs. The proliferating HUVEC cell line and the M2 macrophages demonstrated 10-fold higher expression of the mitotic destabilization gene stathminl (STMN1), as compared to the expression in

Kupffer cells.

An analysis of the cytokine expression profile from M2 macrophages, Kupffer and HUVEC cells infected with YS I646At7.sc// AFLG/ pcigPI ansB/ .csgD strain containing a plasmid encoding huIL- l 5Ra-IL-l 5sc + huSTINGN154S/R284G tazCTT are shown below. Only M2 macrophages expressed IFN-P, IP-10/CXCL10 and I-TAC/CXCL11 in a payload-dependent manner. Kupffer cells, which are highly phagocytic but do not proliferate, do not induce a payload-dependent cytokine profile.

HUVEC cells are not phagocytic, and therefore less permissive to infection.

EXAMPLE 47

Classification of Tumor-Associated Macrophages by Their SPP1 or C1QC Phenotypes Reveals a Higher Association of SPP1 with Poor Overall Survival Than for C1QC

Among the tumor-resident myeloid cells across solid tumor types, macrophages exhibit the most functional diversity. There are at least 12 different macrophage classifying markers, with the three most commonly found across histologies as ISG15, C1QC, and SPP1. ISG15 macrophages, named for the expression of one of many type I IFN-inducible genes, have several markers that resemble the classical Ml macrophages. In contrast, SPP1 and C1QC have markers more attributed to M2 macrophages (Cheng et al., (2021) Cell 184(3):792-809). SPPI TAMs, named for expression of osteopontin (SPP1), display more wound healing and angiogenic phenotypes, as well as the ability to opsonize using osteopontin to enhance phagocytosis (Schack et al., (2009) J Immunol 182(11):6943- 50). C1QC ⁺ TAMs are the most abundant TAMs across tumor types, and are named for one of the three genes, CIQA, Cl QB, and C1QC, of the Clq complement factor that opsonizes and enhances phagocytosis (Fraser etal., (2009) J Immunol 183(10):6175-85). In addition, C1QC upregulation induces expression of the surface receptor MERTK, which provides phagocytosis of apoptotic cells (Zhou et al., (2020) Immunity 52(2):357-373). MERTK was found to be co-expressed with SPP1 in idiopathic pulmonary fibrosis (IPF) and attributed to a highly phagocytic and proliferative macrophage phenotype (Morse J 54(2): 1802441). In colorectal cancer (CRC), s were associated with the lowest overall survival (OS), while were associated with the highest OS (Zhang etal., (2020) Cell 181:442-59). Due to the plasticity of these macrophage subtypes, a bioinformatics analysis was performed to determine the expression of SPPI and C1QC in tumors vs. healthy tissue, as well as their association with tumor histologies.

For the expression of SPPI and C1QC in tumor vs. healthy tissue, gene expression data from The Cancer Genome Atlas (TCGA) was downloaded from Firehose’s normalized matrices and log2 transformed. The clinical data was downloaded from the NCI clinical TCGA database (cancer.gov). Using TCGA gene expression of tumor types with >10 samples, -values were calculated using the Mann Whitney U test. Distributions of SPPI or C1QC expression between tumor and normal samples and the direction determined using the estimate from a generalized linear model were compared. Color of /?-value represents direction; gray /?- value, lower in tumor; light gray /?- value, higher in tumor; very light gray p- value, no significance. The cancer types are ordered by the median expression of the tumor samples.

SPPI was broadly upregulated in tumor tissues vs. normal tissue, and highly upregulated in lung squamous (LUSC) and lung adenocarcinoma (LU AD), breast cancer (BRCA), head and neck squamous cell carcinoma (HNSCC), gastric (stomach) adenocarcinoma (STAD), liver hepatocellular carcinoma (LH4C) and colorectal adenocarcinoma (COAD) (see FIG. 18). Kidney clear cell (KIRC) was the only tumor type that had higher expression of SPPI in normal tissues as compared to tumor tissue. In direct contrast to the SPP1 results, C1QC is not broadly upregulated in tumor tissues, yet is upregulated in kidney tumors, head and neck, esophageal and thyroid (see FIG. 19). In contrast, tumors such as lung, liver and colon had much lower C1QC expression in tumors vs. healthy tissue. The results are set forth in FIGs. 18 and 19.

For the association of SPP1 and C1QC with OS, gene expression from TCGA was used to calculate the Cox proportional hazard regression model, to test for the association between prognosis and expression of SPP1 or C1QC as a continuous variable. The results are displayed as a forest plot, which shows the results for all tumor types with >20 events (see FIGs. 21 and 22, for SPP1 and C1QC respectively). The shape of the dot represents the /?- value, with a triangle representing p < 0.05. The dot is the hazard ratio for each comparison and the black lines are the 95% confidence intervals.

SPP1 was associated with poor survival across many tumor types, with significant associations found in low grade glioma (LGG), cervical squamous carcinoma (CESC), pancreatic adenocarcinoma (PAAD) and liver cancer. Uveal melanoma (UVM) was the only cancer type that had a significant negative association. In contrast to SPP1, C1QC was significantly associated with less tumor types (4 vs. 7), with positively significant associations found in low grade glioma, kidney renal carcinoma (KIRC), and interestingly in uveal melanoma, opposite to that found for SPP1. Melanoma (SKCM) was the only tumor type where C1QC expression was significantly associated with better survival. The results are set forth in FIGs. 21 and 22.

In previous forest plots, SPP1 was associated with poor survival in colon adenocarcinoma (COAD) (p=0.06), while for C1QC, there was no association with overall survival (OS). Colorectal cancer (CRC), which includes colon (COAD) and rectal adenocarcinoma (READ) cancer has been classified into Consensus Molecular Subtypes (CMS) based on myeloid/T-cell ratios and tumor immunophenotypes, and validated across the TCGA, PDX, CMScaller, CRIS and CIRC databases (Guinney et al., (2015) Nat Med. 21(11): 1350-6; Lee etal., (2020) Nature Genetics 52(6):594- 603). CMS4, comprising 23% of CRC patients, has the highest myeloid/T-cell ratio, with a more mesenchymal, TGF0-driven, wound-healing phenotype and the poorest overall survival. CMS1 at 14% of CRC has the next highest myeloid/T-cell ratio, with the most inflamed phenotype and high microsatellite instability (MSI-high), thus a high tumor mutation burden and sensitivity to CPI, and the highest OS. CMS3 at 13% has a dysregulated metabolic phenotype and the lowest myeloid/T-cell ratio. The most common subtype at 37% of CRC is CMS2, with a low myeloid/T-cell ratio and an epithelial phenotype driven by WNT/p-catenin and MYC signaling pathways (Menter el al.., (2019) Curr. Gastroenterol Rep 21(2) 5).

A bioinformatics analysis was performed to assess the association of SPP1 and C1QC with OS, stratifying colon and rectal cancer together within their CMS subtypes. There were too few events (<10) to test CMS3. C1QC was negatively prognostic for CMS1 (p=0.05) and was the only significant association (see KM plot in FIG. 20).

For this analysis, box and whisker plots were generated from SPP1 (see FIG. 41) or C1QC (see FIG. 42) expression across TCGA, and compared to cancer types ordered by median expression. The COAD and READ colorectal cancers were separated into their CMS subtypes and plotted. The distribution of SPP1 expression and C1QC expression in TCGA by cancer type (colorectal cancer split by CMS) was calculated. For both SPP1 and C1QC, CMS4 and 1 had higher protein expression and association with poor OS compared to CMS2 and 3 (see FIGs. 41 and 42). For both SPP1 and C1QC, CMS2 and 3 had low expression and low association with OS compared to the other tumor types as well. If the associations were performed with the SPP1 and C1QC markers stratified into the 4 high and low subtypes, there would be a stronger association with SPP l ^hlgh ClQC ^low and poor OS across CRC and possibly other tumor types (Zhang et al., (2020) Cell 181 :442-59). In summary, SPP1 ⁺ TAMs have a much higher association across solid tumor types with poorer OS compared to C1QC, as well as much higher expression in tumors vs. healthy tissue.

EXAMPLE 48

Infection of Immunomodulatory Bacterial Strains Containing Plasmids Encoding huIL-15Ra-IL-15sc and huSTING N154S/R284G tazCTT In Primary Human M2 Phenotype Macrophages Induces a Hybrid SPP1 ⁺ /C1QC ⁺ Phenotype

An experiment was performed to assess whether primary human monocyte- derived Ml and M2 macrophages express SPP1 and/or C1QC, and to determine whether YS1646Aa«Z/ AFLG/ pagP/ ans RcspI) strain containing a plasmid encoding huIL-15Rci-IL-15sc + huSTING N154S/R284G tazCTT altered the macrophage phenotypes beyond the canonical M1/M2 markers previously described. Frozen human monocytes, isolated from healthy human donors, were thawed in complete medium (RPMI-1640 + 10% FBS), and washed by washed by centrifugation for 10 minutes at 600xg at room temperature. Monocytes were resuspended in ImmunoCult™- SF Macrophage Medium (StemCell Technologies) containing 100 ng/mL human M-CSF. Monocytes (2.5e5 per well) then were seeded in a 48-well plate with a final volume of 500 pL. Five days later (on day 5), fresh medium with cytokines were added to generate Ml and M2 macrophages. To generate primary human Ml macrophages, 250 pL of ImmunoCult™-SF Macrophage Medium containing 10 ng/mL LPS + 50 ng/mL ZFNv was added per well and incubated for 48 hours. To generate primary human M2 macrophages, 250 pL of ImmunoCult™- SF Macrophage Medium containing 150 ng/mL human M-CSF + 60 ng/mL human, IL-4 + 60 ng/mL, and human IL-10 was added per well and incubated for 48 hours.

On day 7, duplicate wells were infected at an MOI of 10, for two hours in RPMI, with the YS1646Aa«//AFLG/A ra /Aan B/Ac D strain containing a plasmid encoding huIL-15Ra-IL-15sc + huSTINGN154S/R284G tazCTT. The cells then were washed 3 times with PBS and resuspended in RPMI + 100 pg/mL gentamicin (Sigma). After 48 hours, the cells were either collected for gene expression analysis using qPCR analysis (as previously mentioned) or cell surface protein expression analysis using flow cytometry.

For qPCR analysis, the cells were collected for RNA isolation using the RNeasy® Plus Mini Kit (Qiagen), according to the manufacturer’s instructions. Total RNA concentration was measured using a NanoDrop™ 2000 UV-Vis Spectrophotometer (Thermo Scientific). The purity of each sample also was assessed from the A260/A230 absorption ratio. Synthesis of cDNA was performed from 0.5-1 pg of template RNA using a CFX96™ Real-Time System (Bio-Rad) and i Script™ Reverse Transcription Supermix for RT-qPCR (Bio-Rad) in a 20 pL reaction, according to the manufacturer’s instructions. qPCR was performed with a CFX96 Real-Time System (Bio-Rad). PrimePCR™ Probe Assay for huCD80 (qHsaCIP0026764), huCD86 (qHsaCEP0050792), huCCR7 (qHsaCIP0033364), huCD14 (qHsaCEP0053971), huSPPl (qHsaCED0057074) and huCIQC (qHsaCED0003470) (All from Bio-Rad). The qPCR reaction (20 pL) was conducted per protocol, using either the SsoAdvanced™ Universal SYBR Green Supermix or the iQ Multiplex Powermix (Bio-Rad). The standard thermocycling program on the Bio-Rad CFX96 Real-Time System consisted of a 95 °C denaturation for 150 sec, followed by 39 cycles of 95 °C for 15 sec and 60 °C for 55 sec. Quantification of the target mRNA was normalized using Actin reference mRNA (Bio-Rad, qHsaCEP0036280). AC _q was calculated as the difference between the target and reference gene.

For flow cytometry analysis, cells were washed with 500-uL of DPBS and detached by incubating with 10-mM EDTA in PBS on ice for 20-min. Lifted cells were washed once with flow buffer (PBS + 2% FBS) and resuspended in 100-uL of LIVE/DEAD Violet Fixable Viability Dye. After 20-min incubation at RT, cells were washed once with flow buffer and resuspended in 50 pL of flow buffer containing True-Stain Monocyte Blocker and Human TruStain FcX. After 10-min incubation on ice, cell surface flow cytometry antibodies CD80 BV605 clone 2D 10 (BioLegend) were added. After 20-min of incubation on ice and in the dark, cells were washed twice with flow buffer. Cells then were resuspended in flow buffer and data were acquired using the NovoCyte® flow cytometer (ACEA Biosciences, Inc.) and analyzed using FlowJo™ software (Tree Star, Inc.).

As shown in the table below, compared to the corresponding untreated Ml or M2 macrophages controls, the YS 1646 A«.sz//AFLG/A/ / /7Ar///.sA Ac.sg'D strain containing a plasmid encoding huIL-15Ra-IL-15sc + huSTINGN154S/R284G tazCTT induced high levels of the Ml macrophage markers CD80, CD86, CCR7 and CD14 on both Ml and M2 macrophages. The Ml macrophages initially had higher expression of C1QC than the M2 macrophages, but infected M2 macrophages showed enhanced expression of SPP1 and C1QC following infection, unlike the Ml macrophages which were not altered. These data demonstrate the ability of the therapy to induce a hybrid SPP1/C1QC phenotype in primary human M2 macrophages.

A follow-up study was performed to further characterize the SPP1/C1QC phenotype in primary human M2 macrophages infected with the YS I646Ac/.s<7 ^/ AFLG/ pagPI ansB/ csgD strain containing a plasmid encoding huIL- I5Ra-IL-15sc + huSTINGN154S/R284GtazCTT, and to assess whether the M2 macrophages can be converted to a hybrid M1/M2 phenotype with enhanced expression of SPP1, C1QC and the ClQC-induced phagocytosis receptor MERTK.

Frozen human monocytes, isolated from healthy human donors, were thawed in complete medium (RPMI-1640 + 10% FBS), and washed by centrifugation for 10 minutes at 600xg at room temperature. Monocytes were resuspended in ImmunoCult™- SF Macrophage Medium (StemCell Technologies) containing 100 ng/mL human M-CSF. Monocytes (2.5e5 per well) then were seeded in a 48-well plate with a final volume of 500 pL. Five days later (on day 5), fresh medium with cytokines were added to generate Ml and M2 macrophages. To generate primary human Ml macrophages, 250 L of ImmunoCult™-SF Macrophage Medium containing 10 ng/mL LPS + 50 ng/mL IFNy was added per well and incubated for 48 hours. To generate primary human M2 macrophages, 250 pL of ImmunoCult™- SF Macrophage Medium containing 150 ng/mL human M-CSF + 60 ng/mL human IL-4 + 60 ng/mL human IL- 10 was added per well and incubated for 48 hours.

On day 7, duplicate wells were infected at an MOI of 10, for two hours in RPMI, with the YS I646Ar/.st// AFLG/ pagPi ansB/ csgD strain containing a plasmid encoding huIL-15Ra-IL-15sc + huSTINGN154S/R284G tazCTT. The cells then were washed 3 times with PBS and resuspended in RPMI + 100 pg/mL gentamicin (Sigma). After 48 hours, the cells were either collected for gene expression analysis using qPCR analysis (as previously mentioned) or cell surface protein expression analysis using flow cytometry.

For qPCR analysis, the cells were collected for RNA isolation using the RNeasy® Plus Mini Kit (Qiagen), according to the manufacturer’s instructions. Total RNA concentration was measured using a NanoDrop™ 2000 UV-Vis Spectrophotometer (Thermo Scientific). The purity of each sample also was assessed from the A260/A230 absorption ratio. Synthesis of cDNA was performed from 0.5-1 pg of template RNA using a CFX96™ Real-Time System (Bio-Rad) and i Script™ Reverse Transcription Supermix for RT-qPCR (Bio-Rad) in a 20 pL reaction, according to the manufacturer’s instructions. qPCR was performed with a CFX96 Real-Time System (Bio-Rad). PrimePCR™ Probe Assay for huSPPl (qHsaCED0057074), huCIQC (qHsaCED0003470), huMYC (qHsaCIP0028650) and huMERTK (qHsaCIP0031506) (All from Bio-Rad). The qPCR reaction (20 pL) was conducted per protocol, using either the SsoAdvanced Universal SYBR® Green Supermix or the iQ Multiplex Powermix (Bio-Rad). The standard thermocycling program on the Bio-Rad CFX96 Real-Time System consisted of a 95 °C denaturation for 150 sec, followed by 39 cycles of 95 °C for 15 sec and 60 °C for 55 sec. Quantification of the target mRNA was normalized using Actin reference mRNA (Bio-Rad, qHsaCEP0036280). AC _q was calculated as the difference between the target and reference gene.

As shown in the table below, compared to the corresponding untreated M2 macrophages controls, the YS1646Aa«Z/ AFLG/ pagP '/ ansB/ csgD strain containing a plasmid encoding huIL-15Ra-IL-15sc + huSTING N154S/R284G tazCTT induced high levels of SPP1 and C1QC expression on macrophages. The STACT infected M2 macrophages demonstrate enhanced expression of the apoptotic cell phagocytosis receptor MERTK, as well as upregulation of the proliferation gene MYC. These data demonstrate that treatment of M2 macrophages with the immunostimulatory bacterial therapy induces a hybrid SPP1/C1QC phenotype with enhanced phagocytic and proliferative capacity.

EXAMPLE 49

Immunomodulatory Bacterial Strains Containing Plasmids Encoding Human huIL-15R«-IL-15sc and huSTING N154S/R284G tazCTT Demonstrate Significant T-Cell and NK Cell Recruitment in MC38 Mouse Model of Colorectal Cancer and Significant T-Cell Recruitment in the KPI Mouse Model of Small Cell Lung Cancer, Activity Which is Correlated to Tumor SPP1/C1QC Expression

This example characterizes the tumor immunophenotype of murine tumor models IV administered the YS1646 asd/ AFLG/ pagPI ansB/l csgD strain containing a plasmid, such as one encoding huIL-15Ra-IL-15sc + huSTING N154S/R284G tazCTT, and assesses whether the ability to deliver payloads correlates with the SPP1/C1QC phenotypes in these models. For this, 5e5 MC38 Cal syngeneic colorectal tumor cells (a gift from the Immunology Department at UC Berkeley) were subcutaneously implanted in the flank of C57BL/6 mice (The Jackson laboratory) and tumor formation was monitored for 10 days. On day 10 post-implantation, mice were intravenously injected with PBS, le7 CFU of N 646Aasd/AFLG/ApagP/AansB/AcsgD containing a control plasmid or le7 CFU of 646Aasd/AFLG/ApagP/AansB/AcsgD containing the plasmid encoding muIL- 15Ra-IL-15sc, huSTING N154S/R284G tazCTT or muIL-15Ra-IL-15sc and huSTING N154S/R284G tazCTT. Seven days after treatment with immunomodulatory bacterial strains, tumors were excised from animals and dissociated using Miltenyi Biotec C tubes containing 200ug/ml Collagenase IV and 20ug/ml DNase I (Sigma- Aldrich). The resulting tumor single cell suspension was stained using anti-murine CD4-FITC conjugated antibodies, anti-murine CD8-BV650 conjugated antibodies, anti -murine NK 1.1 PE conjugated antibodies, Precision Counting Beads (all from Biolegend). In a separate staining, the tumor single cell suspension was stained to assess the expression of PD-1 and PD-L1 on tumor- associated macrophages and tumor cells using anti-murine PD-1 BV785 -conjugated antibodies and anti-murine PD-L1-BV650 conjugated antibodies (all from Biolegend). The data was acquired using the ACEA Novocyte flow-cytometer. The absolute number of CD4 ⁺ and CD8 ⁺ T-cells as well as the absolute number of NK cells was calculated and normalized to the tumor weight and the mean fluorescence intensity was calculated to monitor PD-1 and PD-L1 expressions.

As shown in the table below, recruitment and expansion of CD4 ⁺ and CD8 ⁺ T- cells, as well as Natural Killer (NK) cells, was observed for the muIL-15Ra-IL-15sc payload alone, and enhanced with the combination of muIL-15Ra-IL-15sc + huSTING N154S/R284G tazCTT payloads. PD-1 and PD-Ll expression was detected in tumor-associated macrophages and was increased upon treatment with S1646Aasd/AFLG/ApagP/AansB/AcsgD containing the plasmid encoding the muIL- 15Ra-IL-15sc and huSTING N154S/R284G tazCTT. PD-L1 expression was not substantially detected at the surface of tumor cells and did not significantly increase with treatment.

A similar experiment was performed in the KPI mouse model of immune desert small-cell lung cancer (SCLC). For this, KPI cells (a gift from the Sage lab at Stanford) were sub-cutaneously implanted in the flank of B6129SF1/J mice (The Jackson laboratory) and tumor formation was monitored for 10 days. On day 10 postimplantation, mice were intravenously injected with PBS or le7 CFU of 646Aasd/AFLG/ApagP/AansB/AcsgD containing the plasmid encoding huIL- 15Ra-IL-15sc and huSTING N154S/R284G tazCTT. Twelve days after treatment with immunomodulatory bacterial strains, tumors were excised from animals and dissociated using Miltenyi Biotec C tubes containing 2000ug/ml Collagenase IV and 20ug/ml DNase I (Sigma- Aldrich). The resulting tumor single cell suspension was stained using anti-murine CD4-FITC conjugated antibodies, anti-murine CD8-BV650 conjugated antibodies, Precision Counting Beads (all from Biolegend) and the data was acquired using the ACEANovocyte flow-cytometer. The absolute number of CD4 ⁺ and CD8 ⁺ T cells was calculated and normalized to the tumor weight.

The immunostimulatory bacteria encoding huIL-15Ra-IL-15sc and huSTING N154S/R284G tazCTT induced a significant recruitment of CD4 ⁺ and CD8 ⁺ T in the tumor microenvironment as compared to the PBS-treated group. While no tumor regressions were observed, likely due to the lack of tumor antigens in this immune desert model, it demonstrated the on-target ability of YS1646Aasd/AFLG/ApagP/AansB/AcsgD containing the plasmid encoding the huIL- 15Ra-IL-15sc and huSTING N154S/R284G tazCTT immunomodulatory payloads to deliver type I IFN and huIL-15Ra-IL-15sc to recruit and expand T-cells in these tumors.

Next an experiment was performed with the tumors from the above MC38 Cal and KPI tumor models, in which the immunostimulatory bacteria were efficacious, and compared to tumors previously isolated from a MC38 (Kerafast) tumor model that did not demonstrate T-cell infiltration or anti-tumor efficacy, in order to determine whether tumor SPP1/C1QC status may be correlated to in vivo payload delivery.

On day 7 post-IV injection, tumors were excised and processed for RNA extraction using the GentleMACS™ Octo Dissociator and the C tubes (Miltenyi Biotec) molecule setting in 2 mL of PBS. The homogenates were spun down at 1300 RPM for 10 minutes, and single cells suspension was collected for RNA isolation using the RNeasy® Plus Mini Kit (Qiagen), according to the manufacturer’s instructions. Total RNA concentration was measured using a NanoDrop™ 2000 UV- Vis Spectrophotometer (Thermo Scientific). Previously stored RNA from the MC38 Kerafast model was also used. The purity of each sample also was assessed from the A260/A230 absorption ratio. Synthesis of cDNA was performed from 0.5-1 pg of template RNA using a CFX96™ Real-Time System (Bio-Rad) and iScript™ Reverse Transcription Supermix for RT-qPCR (Bio-Rad) in a 20 pL reaction, according to the manufacturer’s instructions. qPCR was performed with a CFX96 Real-Time System (Bio-Rad). Custom qPCR primers were used for eSTING detection. PrimePCR™ Probe Assay for muSppl (qMmuCEP0062824), muClqc (qMmuCEP0057437) were purchased from Bio-Rad. The qPCR reaction (20 pL) was conducted per protocol, using either the SsoAdvanced Universal SYBR® Green Supermix or the iQ Multiplex Powermix (Bio-Rad). The standard thermocycling program on the Bio-Rad CFX96 Real-Time System consisted of a 95 °C denaturation for 150 sec, followed by 39 cycles of 95 °C for 15 sec and 60 °C for 55 sec. Quantification of the target mRNA was normalized using Actin reference mRNA (Bio-Rad, qHsaCEP0036280). AC _q was calculated as the difference between the target and reference gene. The values are shown in the table below as the average of four mice.

Compared to the PBS control group, 6 6Aasd/AFLG/ApagP/AansB/AcsgD strains containing the plasmid encoding the huIL-15Ra-IL-15sc and huSTING N154S/R284G tazCTT induced high levels of huSTING N154S/R284G tazCTT expression in KPI and MC38 Cal tumors, but not in the tumors from the MC38 Kerafast model. PBS-treated tumors from the three models were assessed in order to determine whether Sppl/Clqc status may be predictive of payload delivery and efficacy. As shown below, PBS-treated tumors from the MC38 Kerafast model expressed high levels of Sppl and low levels of Clqc, whereas for the other two models it was the opposite. These data imply that a high Sppl/ClqC ratio in a colorectal tumor model may negatively impact the ability to deliver plasmids.

EXAMPLE 50

Immunostimulatory Bacteria encoding IL-15Ra-IL15sc + huSTING N154S/R284G tazCTT Demonstrates Potent Payload Delivery in Orthotopic EMT6 Tumors and Immune Correlates of Efficacy

A study was performed to evaluate the in vivo immune correlates of potency in an orthotopic, T-cell excluded, and metastatic model of CPI (immune checkpoint inhibitor) refractory triple-negative breast cancer. For this experiment, 6-8 week-old female BALB/c mice (10 mice per group for TGI, 5 mice per group for immune correlates) were inoculated in the left mammary fat pad with EMT6 tumor cells (ATCC® CRL-2755™) (IxlO ⁶ cells in 100 pL PBS). Mice bearing 7 day-old established mammary tumors (~65 mm ³ in volume) were intravenously (IV) injected with a single dose of 3 xlO ⁷ CFUs of YS I 646Aa.st// AFLG/ ApagP! ansB! csgL) strains contained plasmids encoding muIL-15Ra-IL15sc + huSTING N154S/R284G tazCTT, or huIL-15Ra-IL15sc + huSTINGN154S/R284G tazCTT and were compared to treatment with PBS control.

The tumors in the PBS-treated mice grew evenly, reaching a max tumor volume at day 31. Mice that were IV treated with either the bacterial strain containing a plasmid encoding muIL-15Ra-IL15sc + huSTINGN154S/R284G tazCTT or with huIL-15Ra-IL15sc + huSTING N154S/R284G tazCTT, both achieved a 70% complete response rate. The therapeutic payload combinations were very well tolerated, and mice did not lose weight during the study. These data demonstrate the in vivo potency of IV administered immunostimulatory bacteria encoding the combinations of muIL-15Ra-IL15sc + huSTINGN154S/R284G tazCTT, and huIL- I5Ra-ILI5sc + huSTING N154S/R284G tazCTT in an orthotopic, T-cell excluded, metastatic model of checkpoint inhibitor refractory triple-negative breast cancer (TNBC).

In the separately treated cohort (N=5), mice were harvested at D7 in order to interrogate tumor immune correlates of potency. In the group treated with huIL-15Ra- IL-15sc + huSTINGN154S/R284G tazCTT, 3 out of 5 mice had already cured out and only 2 mice remained for analysis. All 5 tumors, while small, remained for analysis in the muIL-15Ra-IL-15sc + huSTINGN154S/R284G tazCTT group. On day 7 post-IV injection, tumors were assessed for expression of muIL-15Ra-IL-15sc or huIL-15Ra-IL-15sc and huSTINGN154S/R284G tazCTT, relative to PBS control. Excised tumors were processed for RNA extraction using the GentleMACS™ Octo Dissociator and the C tubes (Miltenyi Biotec) molecule setting in 2 m of PBS. The homogenates were spun down at 1300 RPM for 10 minutes, and single cells suspension was collected for RNA isolation using the RNeasy® Plus Mini Kit (Qiagen), according to the manufacturer’s instructions. Total RNA concentration was measured using a NanoDrop™ 2000 UV-Vis Spectrophotometer (Thermo Scientific). The purity of each sample also was assessed from the A260/A230 absorption ratio. Synthesis of cDNA was performed from 0.5-1 pg of template RNA using a CFX96™ Real-Time System (Bio-Rad) and iScript™ Reverse Transcription Supermix for RT- qPCR (Bio-Rad) in a 20 pL reaction, according to the manufacturer’s instructions. qPCR was performed with a CFX96 Real-Time System (Bio-Rad). Custom qPCR primers were used for eSTING and IL-15 detection. PrimePCR™ Probe Assay for mulfnci2 (qMmuCEP0043629), mulfnpi (qMmuCEP0058870), muCxcllO (qMmuCEP0057926) and mu4-lbbl (qMmuCED0004214), muSppl (qMmuCEP0062824), muClqc (qMmuCEP0057437) were purchased from Bio-Rad. The qPCR reaction (20 pL) was conducted per protocol, using either the SsoAdvanced Universal SYBR® Green Supermix or the iQ Multiplex Powermix (Bio-Rad). The standard thermocycling program on the Bio-Rad CFX96 Real-Time System consisted of a 95 °C denaturation for 150 sec, followed by 39 cycles of 95 °C for 15 sec and 60 °C for 55 sec. Quantification of the target mRNA was normalized using Actin reference mRNA (Bio-Rad, qHsaCEP0036280). AAC _q was calculated as the difference between the treatments and the uninfected control.

The values are shown in the table below as the average of 5 mice for the muIL-15Ra-IL-15sc + huSTINGN154S/R284G tazCTT group, and 2 mice for huIL- 15Ra-IL-15sc + huSTING N154S/R284G tazCTT group, relative to an average of 5 mice in the PBS control group. Compared to the PBS control group, both tumor groups treated with the strains containing muIL-15Ra-IL-15sc or huIL-15Ra-IL-15sc + huSTING N154S/R284G tazCTT induced high levels of expression in tumors of their respective payloads, as compared to PBS. Both treatment groups induced high levels of payload-dependent Ifna2, Ifnpi, CxcllO, and the costimulatory receptor 4- Ibbl. In this model, the PBS group had a mixed Sppl/Clqc phenotype, with higher expression of Sppl relative to Clqc. Both treatment groups induced C1QC expression, and to a higher degree in the huIL-15Ra-IL-15sc + huSTING N154S/R284G tazCTT treatment group, and both demonstrated reduced SPP1 expression relative to the PBS group.

These data describe the tumor genetic signatures of potency for strains delivering huIL-15Ra-IL-15sc + huSTING N154S/R284G tazCTT payloads, as well as an immune correlate in breast cancer models that upregulation of C1QC, and downregulation of SPP1 in tumor-resident macrophages may be an indicator of potency and payload delivery.

EXAMPLE 51

Immunomodulatory Bacterial Strains, Containing Plasmids Encoding Human IL-15Ra-IL-15sc and huSTING N154S/R284G tazCTT Demonstrate In Vivo Payload Delivery, T-Cell Recruitment and Efficacy, Correlating with High Tumor Proliferating Macrophages

The goal of this experiment is to demonstrate optimal tumor immunophenotype for treatment with the F\ M6Aasd AFIdi ApagP Aaii B Acsgl) strain containing the plasmid encoding huIL-15Ra-IL-15sc and huSTING N154S/R284G tazCTT to induce payload delivery and anti -tumor immunity in syngeneic tumor mouse models.

The MC38 model (colorectal), EMT6 model (breast), CT26 (colon) and 4T1 (breast) models of cancer were used in this example. MC38 and CT26 cells were subcutaneously implanted in the flank of C57BL/6 and Balb/c mice (The Jackson laboratory) respectively, and tumor formation was monitored for 10 days. 4T1 and EMT-6 cells were orthotopically implanted in the mammary pad of Balb/c mice and tumor formation was monitored for 10 days.

On day 10 post-implantation (day 0), tumors were excised from animals prior to treatment and dissociated using Miltenyi Biotec C tubes containing 200ug/ml Collagenase IV and 20ug/ml DNase I (Sigma-Aldrich). The resulting tumor single cell suspension was stained using anti-murine CD8-BV650, anti-murine F4/80 APC, anti-murine Ki67 BV421 conjugated antibodies and Precision Counting Beads (all from Biolegend). Mice were treated with PBS or 6 6Aasd/AFLG/ApcigP/AansB/AcsgD strain containing the plasmid encoding either Nano-luciferase (used as a control strain) or huIL-15Ra-IL-15sc + huSTING N154S/R284G tazCTT. Seven days after treatment with immunomodulatory bacterial strains, the tumors were dissociated as described earlier and single cell suspensions were stained using anti-murine CD8-BV650 conjugated antibodies and Precision Counting Beads (all from Biolegend). The data was acquired using the ACEA Novocyte flow-cytometer. The absolute number of CD8+ T cells at day 0 and day 7 as well as the absolute number of tumor-associated Ki67+ proliferating macrophages at day 0 was calculated and normalized to the tumor weight for every syngeneic models.

The level of tumor Mertk, Sppl and Clqc expression was assessed to compare the tumor-specific gene signature in various tumor types. On day 0 prior to IV injection, tumors were excised and processed for RNA extraction using the GentleMACS™ Octo Dissociator and the C tubes (Miltenyi Biotec) molecule setting in 2 mL of PBS. The homogenates were spun down at 1300 RPM for 10 minutes, and single cells suspension was collected for RNA isolation using the RNeasy® Plus Mini Kit (Qiagen), according to the manufacturer’s instructions. Total RNA concentration was measured using a NanoDrop™ 2000 UV-Vis Spectrophotometer (Thermo Scientific). The purity of each sample also was assessed from the A260/A230 absorption ratio. Synthesis of cDNA was performed from 0.5-1 pg of template RNA using a CFX96™ Real-Time System (Bio-Rad) and iScript™ Reverse Transcription Supermix for RT-qPCR (Bio-Rad) in a 20 pL reaction, according to the manufacturer’s instructions. qPCR was performed with a CFX96 Real-Time System (Bio-Rad). Custom qPCR primers were used for eSTING detection. PrimePCR™ Probe Assay for muMertk (qMmuCEP0052884), muSppl (qMmuCEP0062824) and muClqc (qMmuCEP0057437) were purchased from Bio-Rad. The qPCR reaction (20 pL) was conducted per protocol, using either the SsoAdvanced Universal SYBR® Green Supermix or the iQ Multiplex Powermix (Bio-Rad). The standard thermocycling program on the Bio-Rad CFX96 Real-Time System consisted of a 95 °C denaturation for 150 sec, followed by 39 cycles of 95 °C for 15 sec and 60 °C for 55 sec. Quantification of the target mRNA was normalized using Actin reference mRNA (Bio-Rad, qHsaCEP0036280). AC _q was calculated as the difference between the target and reference gene.

As shown in the tables below, strong anti-tumor activity after treatment with immunomodulatory bacterial strains was observed for the syngeneic models that were highly enriched for proliferating tumor-associated macrophages (MC38 and EMT6). Tumor growth inhibition was not observed in the syngeneic models containing low levels of proliferating tumor-associated macrophages (4T1 and CT26). Similarly, increases in CD8 ⁺ T cell recruitment also correlated with proliferating macrophages, and were highest in the MC38 model, followed by the EMT6 model. For the tumor qPCR measuring Sppl and Clqc, the 4T1 and CT26 models had expression of Sppl and of Clqc at similar levels, whereas EMT6 had lower expression of both. MC38 had the highest Clqc, very low Sppl, and was the only model that demonstrated expression of Mertk. When comparing the ratio of Sppl to Clqc to the tumor efficacy, the highest ratios correlate with the lowest efficacy. These data demonstrate that tumor composition prior to dosing can be predictive of payload delivery and anti-tumor efficacy. The higher the amount of proliferating macrophages, as well as the lower the Sppl to Clqc ratio, the more payload delivery will occur and result in type I IFN-mediated CD8 ⁺ T cell recruitment and anti-tumor immunity upon treatment with STACT-encoding huIL-15Ra-ZL-15sc + huSTING N154S/R284G tazCTT.

EXAMPLE 52

Immunomodulatory Bacterial Strains, Containing Plasmids Encoding Human IL-15R«-IL-15sc and huSTING N154S/R284G tazCTT Demonstrates Significant T-Cell Recruitment in a Spontaneous Autochthonous Mouse Model of Breast Cancer

The spontaneous model of breast cancer, MMTV-PyMT, was employed to further illustrate the on-target potency and immune correlates of theYS1646Aasz//AFLG/ pagPI ans ! csgl) strain containing a plasmid encoding huIL-15Ra-IL-15sc + huSTING N154S/R284G tazCTT in a spontaneous and immune desert model of breast cancer. In this model, females develop palpable mammary tumors with near 100% penetrance due to the expression of the Polyoma Virus middle T antigen under the direction of the mouse mammary tumor virus promoter/enhancer.

For this study, mice were IV injected with 3 xlO ⁷ CFUs of the bacterial strain or PBS control when their largest tumors measured an average of 272 mm ³ in volume (between day 79 to 96 post-acquisition of the mice). Tumors were collected between days 85 to 121 post-acquisition of the mice. For bacterial enumeration, Tumors were homogenized Miltenyi Biotec M tubes and plated on LB plates to enumerate the number of CFUs per gram of tumor tissue. All tumors collected were found to be well colonized despite a range of tumor weights (0.05 g to 0.36 g, N=7), with a mean of 2.89 x 10 ⁶ CFUs/g. These data demonstrate comparable tumor colonization of spontaneous MMTV-PyMT across all tumors tested. For flow cytometry, tumors were dissociated using Miltenyi Biotec C tubes containing 200ug/ml Collagenase IV and 20ug/ml DNase I (Sigma- Aldrich). The resulting tumor single cell suspensions were stained using anti-murine CD4-FITC conjugated antibodies, anti-murine CD8-BV650 conjugated antibodies, anti-murine F4/80 APC, anti-murine Ki67 BV421 (all from Biolegend) and the data was acquired using the ACEA Novocyte flow-cytometer. The absolute number of CD4 ⁺ T cells and CD8 ⁺ T cells as well as the total amount of tumor-associated macrophages and the sub-population of proliferating macrophages was calculated. This Example describes the impact of the delivery of various immunomodulatory payload by the immunostimulatory bacteria in mammary tumors. The immunostimulatory bacteria encoding huIL-15Ra-IL-15sc + huSTINGN154S/R284G tazCTT induce a significant recruitment of CD4 ⁺ and CD8 ⁺ T in the tumor microenvironment as compared to the PBS-treated group. The immunostimulatory bacteria encoding huIL-15Ra-IL-15sc + huSTING N154S/R284G tazCTT also increases the number of proliferating macrophages and a positive correlation between the amount of proliferating tumor- associated macrophages and the CD8 ⁺ T cells was observed for the groups treated with the immunostimulatory bacteria encoding huZL-15Rot-IL-15sc + huSTING N154S/R284G tazCTT (p=0.0007; R=0.6674).

For tumor qPCR analysis, a piece each tumor was excised and processed for RNA extraction using the GentleMACS™ Octo Dissociator and the C tubes (Miltenyi Biotec) molecule setting in 2 mL of PBS. The homogenates were spun down at 1300 RPM for 10 minutes, and single cells suspension was collected for RNA isolation using the RNeasy® Plus Mini Kit (Qiagen), according to the manufacturer’s instructions. Total RNA concentration was measured using a NanoDrop™ 2000 UV- Vis Spectrophotometer (Thermo Scientific). The purity of each sample also was assessed from the A260/A230 absorption ratio. Synthesis of cDNA was performed from 0.5-1 pg of template RNA using a CFX96™ Real-Time System (Bio-Rad) and i Script™ Reverse Transcription Supermix for RT-qPCR (Bio-Rad) in a 20 pL reaction, according to the manufacturer’s instructions. qPCR was performed with a CFX96 Real-Time System (Bio-Rad). Custom qPCR primers were used for eSTING detection. PrimePCR™ Probe Assay for muSppl (qMmuCEP0062824), muClqc (qMmuCEP0057437), muMertk (qMmuCEP0052884), and muCxcllO (qMmuCEP0057926) were purchased from Bio-Rad. The qPCR reaction (20 pL) was conducted per protocol, using either the SsoAdvanced Universal SYBR® Green Supermix or the iQ Multiplex Powermix (Bio-Rad). The standard thermocycling program on the Bio-Rad CFX96 Real-Time System consisted of a 95 °C denaturation for 150 sec, followed by 39 cycles of 95 °C for 15 sec and 60 °C for 55 sec. Quantification of the target mRNA was normalized using Actin reference mRNA (Bio-Rad, qHsaCEP0036280). AC _q was calculated as the difference between the target and reference gene.

As shown in the table below, the MMTV-PyMT tumors are SPP I ¹¹¹ , similar to previous breast cancer models, and treatment with YS1646Aasd/AFLG/ApagP/AansB/AcsgD strain containing the plasmid encoding huIL-15Ra-IL-15sc + huSTINGN154S/R284G tazCTT induced higher Clqc and lower Sppl compared to PBS. Further, upregulation of CXCL10 and the phagocytic receptor Mertk are also on target as evidence of successful induction of type I IFN.

These data further demonstrate the mechanism of action of the N 6 6Aasd/AFLG/ApagP/AansB/AcsgD strain containing the plasmid encoding huIL-15Ra-IL-15sc + huSTINGN154S/R284G tazCTT, whereby immune desert tumors with high proliferating macrophage content are well colonized by the bacteria and deliver type I IFN-inducing payloads that promote T-cell recruitment, increases in proliferating macrophages, type I IFN induced CxcllO, as well as increased Clqc and Mertk, and decreased Sppl

EXAMPLE 53

Prospective Biomarkers for Identifying Patients Enriched in Proliferating Macrophages

The presence of proliferating macrophages in solid tumors has been previously described. Biopsies from breast cancer patients at two separate hospitals, UCSF and University of Chicago, were assessed for tumor grade and estrogen receptor (ER), progesterone receptor (PR), or HER2/Erbb2 receptor status. Biopsies were assessed by immunohistochemistry (IHC) for the number of CD68 ⁺ macrophages that also stained for the proliferation marker Proliferating Cell Nuclear Antigen (PCNA). Results between the two hospitals were compared, and both found that CD68 PCNA proliferating macrophages were more abundant in ER/PR negative luminal breast cancer, and basal, triple negative breast cancer (TNBC). Further, proliferating macrophages increased with increasing tumor stage (Campbell etal, (2016) Breast Cancer Res Treat 155:431-440). Interestingly, the authors summarized their findings, yet did not express any theories as to why proliferating macrophages, as opposed to other immunosuppressive TAM phenotypes, may be responsible for the more aggressive breast cancer progression.

In summary, performing routine IHC on patient biopsies and staining for CD68 ⁺ and PCNA ⁺ , or the common proliferation marker Ki67 ⁺ , is a prospective biomarker for tumors across histologies that are likely to respond favorably to immunostimulatory bacterial therapies bearing DNA plasmids, and other therapies delivering DNA to immune cells of the tumor microenvironment.

Beyond IHC biomarkers, assessing biopsies for expression of genes associated with proliferating macrophages also can identify subjects with cancers amenable to delivery of gene therapy to proliferating immune cells. In order to determine common biomarkers across tumor histologies, TCGA databases and published scRNA-seq datasets using lOx and smart-seq technologies were analyzed across solid tumor types. To assess the quantity of proliferating macrophages across histologies, an analysis was performed on tumor types using published scRNA-seq datasets using lOx and smart-seq technologies. Proliferating macrophages were assessed using the G2/M score, which is a set of genes involved in the transition from G2 to M phase of cell cycle, with the distribution defined using the CellCycleScoring function and the Seurat object (Zillionis et al., (2019) Immunity 50(5): 1317-1334). Cells defined as proliferating had a G2M score >14. The proportions are depicted in the bar graph as percentages (0.05 = 5%) and calculated using proportion of CD68 ⁺ cells of total CD45 ⁺ cells, and proportion of G2/M score ⁺ of total CD68 ⁺ macrophages from tumor tissue samples of all patients in each dataset (see FIGs. 23A and B).

Proliferating macrophages were highest overall in the datasets acquired from lung, prostate, and colon cancer. Kidney clear cell cancer has high macrophages, but lowest proliferating macrophages (Obradovic etal., (2021) Cell 184(11):2988-3005). HBV ⁺ Liver cancer has a moderate amount of proliferating macrophages (Zhang et al., (2019) Cell 179(4): 829-845). Head and neck cancer (HNSCC) has the fewest macrophages, likely due to the high TMB from carcinogen or HPV exposure, but of these there are proliferating macrophages (Cillo et al, (2020) Immunity 52(1): 183- 199). Breast cancer has high proliferating macrophages, and these are increased post treatment proportionally with increased macrophages, and also increased following anti-PD-1 (pembrolizumab) therapy (Bassez et al., (2021) Nat Med 27(5):820-832). Prostate cancer is high overall in proliferating macrophages (Chen et al., (2021) Nat Cell Biol 23(l):87-98). Colon cancer also has high proliferating macrophages, and even more in the CMS1 and CMS4 subtypes (Zhang etal., (2020) Cell 181 :442-59; Lee et al., (2020) Nat Genetics 52(6):594-603). Lung cancer (NSCLC) has the highest amount of proliferating macrophages (Zillionis et al., (2019) Immunity 50(5): 1317- 1334). Thus, across solid tumor types, proliferating macrophages can be identified by gene analysis of CD68 and the G2/M gene module. Proportion of macrophages across all datasets is set forth in FIGS. 23A and B.

As apoptotic tumor cells are known to recruit macrophages and efferocytosis of these cells enhances macrophage proliferation (Mortimer etal., (1999) Gene Therapy 6:403-411; Elliott etal., (2009) Nature 461(7261):282-6), a bioinformatics study was performed to assess the association of an apoptosis gene module with solid tumor types, and with the stratification of colon and rectal cancer into their CMS subtypes. For this analysis, box and whisker plots were generated from the apoptosis module expression data from TCGA (161 genes associated with apoptosis), compared to cancer types and ordered by median expression. Boxes represent median ± interquartile range and whiskers ±1.5 x interquartile range. Outliers are represented by black dots. The results are set forth in FIG. 24.

The apoptosis module was highly associated with tumor types previously established as having high macrophage content, such as the colon CMS4 and CMS1 subtypes, lung cancer (LU D, LUSC), and breast cancer. Other potential tumor types identified from this analysis may also predict tumors high in proliferating macrophages, such as mesothelioma, pancreatic cancer, and gastric cancer (STAD). Interestingly, HPV-driven cancers such as head and neck (HNSC) and cervical cancer (CESC) also scored high in the apoptosis module, and the spread in the data may indicate the difference between HPV ⁺ vs. HPV cancers.

Individual patients can be responsive to the immunostimulatory bacteria provided herein, despite having a less favorable histology overall, and can be identified using other gene expression markers. A broad set of gene expression markers associated with proliferation were assessed from all macrophages in the tumor tissue, compiled from the previous list of RNASeq and 10X genomics datasets from the publications above including SPP1, C1QC and the G2M score, and individual genes from the G2M score set. Analyses were performed on lung, breast and colon (no CMC subtyping) datasets from the publications used in the proliferating macrophage analysis above (Zillionis et al., (2019) Immunity 50(5): 1317-1334; Bassez etal., (2021) Afet/27(5):820-832; Zhang etal., (2020) Cell 181:442-59).

As shown in FIGS. 25A-F, proliferating macrophages from different histologies had varying associations with SPP1 or C1QC. For example, in the Zillionis NSCLC lung cancer paper, a subset of proliferating macrophages was identified using the G2M module. Further analysis of the macrophage subtypes revealed that the proliferating macrophage subtype identified in the paper (tMacCycl), had higher SPP1 expression than C1QC and scored very high for the microtubule destabilizer gene stathminl (STMN1). Conversely, pre-treatment biopsies from breast cancer patients were a hybrid of SPP1 and C1QC, while the colon dataset (not stratified into CMS subtypes) revealed that proliferating macrophages were almost entirely C1QC ⁺ . Thus, stratification of patients using tumor expression of SPP1 or C1QC and the G2M score may reveal SPP1 ⁺ or C1QC ⁺ proliferating macrophages in a histology-specific manner. While SPP1 may be predictive for proliferating macrophages in lung cancer and C1QC for colon cancer, other histologies such as breast may be more nuanced.

Across all the histologies analyzed, the two markers most associated with proliferating macrophages were found to be the G2/M score and STMN1. Shown in the box plots in FIGS. 25A-F are these results for the lung, breast and colon data sets, demonstrating that G2/M score and STMN1 are highly predictive of proliferating macrophages. The results are set forth in FIGS. 25A-F.

The correlation between SPP1 and C1QC to somatic genetic tumor mutations was assessed, in order to further identify patients in different histologies who would be selected as responsive to the therapy. For this, association plots were constructed that tested the association of SPP1 expression with alterations (single nucleotide variants, indels, copy number changes, fusions/rearrangements) in the major cancer pathways (Supplementary table 4, Sanchez-Vega et al., (2018) Cell 173(2):321-337) in TCGA. The y-axis represents the -logl0( ?-value), with the higher the dot the lower the -value. The different dots represents different tumor types. The association was tested using logistic regression. Only tumor types and mutation/pathway combinations with >15 samples in both wild-type and altered were tested. Association with genetic mutations can identify patients, who will respond to treatment, where expression of SPP1 or C1QC was weakly associated with proliferating macrophages and expression in tumor vs. healthy tissue, but the range for the association was very broad.

As shown in the SPP1 association plot in FIG. 26, NRF2 pathway alterations were associated with SPP1 expression in squamous carcinomas such as lung (LUSC), head and neck (HNSC), and cervical (CESC), as well as esophageal (ESCA), bladder (BLCA), and kidney renal papillary (KIRP). TP53 mutations in breast cancer (BRCA) and SPP1 were strongly associated. SPP1 and PI3K mutations were identified in prostate cancer (PRAD), and BRAF mutations in skin cutaneous melanoma (SKCM). For the C1QC association plot in FIG. 27, mutations in the HIPPO pathway were associated with C1QC ⁺ proliferating macrophages in uterine corpus endometrial cancer (UCEC), and KMT2A mutations in bladder cancer (BLCA). Interestingly, TP53 pathway mutations were also strongly associated with C1QC in breast cancer (BRCA). This further supports the finding from the box plot analysis above, that proliferating macrophages in breast cancer may be associated with hybrid SPP1 ⁺ C1QC ⁺ , and TP53 pathway mutations and could be an excellent prognostic biomarker. These results are set forth in FIGS. 26 and 27.

In summary, while the correlation of SPP1 and C1QC with proliferating macrophages can be different between and among histologies, determining the histologies that are high in proliferating macrophages can be accomplished by IHC and genetic markers in order to better stratify patients more likely to respond to treatment.

EXAMPLE 54

Treatment of Macrophages with Culture Conditions That Promote Cell Cycle Progression Provides Payload Delivery Following Infection of Immunomodulatory Bacterial Strains Containing Plasmid Payloads Ectopic gene expression following plasmid transfer requires actively proliferating cells (Mortimer etal., (1999) Gene Therapy 6:403-411). While Ml- differentiating factors such as IFNy and LPS have been shown to impair cell cycle progression in macrophages, M2-differentiating factors such as M-CSF and efferocytosis of apoptotic cells were shown to promote macrophage cell cycle progression (Rbszer et al., (2018) Cells 7(8): 103; Nasser et al., (2020) Cell Death Discov 6:63). As differentiation of M2 TAMs is commonly performed using M-CSF with the addition of IL-4 and IL- 10, the role of these cytokines in promoting cell cycle remained to be determined. According to the literature, primary human macrophages could be induced to proliferate with M-CSF, but proliferation was impaired with the addition of GM-CSF or IL-4 (Nasser etal., (2020) Cell Death Discov 6:63). Further, it was shown in mouse and human macrophages that the addition of M-CSF with either IL-4 and/or IL- 10 induced upregulation of MAFB. The role these macrophage differentiation conditions have on promoting macrophage proliferation and payload delivery in our system remained to be determined.

Experiments were performed to assess the effects of adding IL-4 and IL- 10 to the M-CSF differentiation media, as well as whether the addition of apoptotic tumor cells would enhance proliferation and payload delivery.

For this, frozen human monocytes, isolated from healthy human donors, were thawed in complete medium (RPMI-1640 + 10% FBS), and washed by washed by centrifugation for 10 minutes at 600xg at room temperature. Monocytes were resuspended in ImmunoCult™- SF Macrophage Medium (StemCell Technologies) containing 100 ng/mL human M-CSF. Monocytes (2.5e5 per well) then were seeded in a 48-well plate with a final volume of 500 pL. Five days later (on day 5), fresh medium with cytokines were added to generate M0 and M2 macrophages. To generate primary human M0 macrophages, 250 pL of ImmunoCult™-SF Macrophage Medium containing 150 ng/mL human M-CSF was added per well and incubated for 48 hours. To generate primary human M2 macrophages, 250 pL of ImmunoCult™- SF Macrophage Medium containing 150 ng/mL human M-CSF + 60 ng/mL human IL-4 + 60 ng/mL human IL- 10 was added per well and incubated for 48 hours. To promote cell cycle progression, apoptotic Jurkat cells were added to the corresponding media.

On day 7, duplicate wells were infected at an MOI of 10, for two hours in RPMI, with the YS I646A«.sz// AFLG/ pagPI .ansB/ csgD strains containing either a plasmid encoding NanoLuciferase® or huIL-15Ra-IL-15sc + huSTING N154S/R284G tazCTT. The cells were washed 3 times with PBS and resuspended in RPMI + 100 pg/mL gentamicin (Sigma).

After 48 hours, supernatants were harvested and evaluated for downstream signaling differences using a custom human M2/M1 U-Plex® panel (Meso Scale Discovery), according to the manufacturer’s protocol. Cytokine secretion was measured, and the average of the duplicate was calculated. For qPCR analysis, the cells were collected for RNA isolation using the RNeasy® Plus Mini Kit (Qiagen), according to the manufacturer’s instructions. Total RNA concentration was measured using a NanoDrop™ 2000 UV-Vis Spectrophotometer (Thermo Scientific). The purity of each sample also was assessed from the A260/A230 absorption ratio. Synthesis of cDNA was performed from 0.5-1 pg of template RNA using a CFX96™ Real-Time System (Bio-Rad) and i Script™ Reverse Transcription Supermix for RT-qPCR (Bio-Rad) in a 20 pL reaction, according to the manufacturer’s instructions. qPCR was performed with a CFX96 Real-Time System (Bio-Rad). Custom qPCR primers for IL-15, SEQ ID NOs: 496- 499, as above, (For: 5’-GCAAGAGGAATTTGCCATCG (SEQ ID NO:496); Rev: 5’- TTGCTCCATCCCGCTAATG (SEQ ID NO:497)) and STING (For: AACTCTGGTTTCAAGCGTAAAG (SEQ ID NO 498); Rev: CAGGGCAGGGTCTCTAATG (SEQ ID NO:499)) were prepared. The qPCR reaction (20 pL) was conducted per protocol, using either the SsoAdvanced Universal SYBR® Green Supermix or the iQ Multiplex Powermix (Bio-Rad). The standard thermocycling program on the Bio-Rad CFX96 Real-Time System consisted of a 95 °C denaturation for 150 sec, followed by 39 cycles of 95 °C for 15 sec and 60 °C for 55 sec. Quantification of the target mRNA was normalized using Actin reference mRNA (Bio-Rad, qHsaCEP0036280). C _q was calculated as the difference between the target and reference gene.

As shown in the tables below, the presence of apoptotic Jurkat cells induced higher payload delivery in M2 macrophages differentiated in M-CSF, both in gene expression as well as enhanced IFNp over the untransfected and plasmid control groups. The addition of IL-4 and IL-10 to the M-CSF media suppressed payload expression and eliminated the production of IFNP, CXCL10, and payload-dependent IL-16 from STING activity. Interestingly, addition of apoptotic Jurkats partially rescued this phenotype, with higher gene expression, and slightly higher payloadspecific IFNP and IL-6, but no CXCL10.

These data demonstrate that conditions which enhance macrophage cell cycle progression, such as M-CSF and apoptotic cells, enhances payload delivery and payload-driven cytokine production. However, the addition of factors such as IL-4 and IL-10 that impair cell cycle will also impair payload delivery. As MAFB upregulation was previously shown to impair type I IFN production in human macrophages following cytosolic RNA sensing, there may be suppression happening from IL-4 and IL- 10 beyond cell cycle that may impair payload-induced IFNP (Kim et al., (2010) Nat Immunol. 11(8):743-50). In a previous study, it was determined that liver Kupffer cells, while phagocytic, were not proliferating and thus could not transfer and express payloads. Whether these cells could be induced to proliferate by cell culture conditions remained to be determined. To assess this, a study was also performed to determine whether non-proliferating Kupffer cells can be induced to proliferate by culturing with apoptotic hepatocytes.

For this, frozen human monocytes, isolated from healthy human donors, were thawed in complete medium (RPML1640 + 10% FBS), and washed by washed by centrifugation for 10 minutes at 600xg at room temperature. Monocytes were resuspended in ImmunoCult™- SF Macrophage Medium (StemCell Technologies) containing 100 ng/mL human M-CSF. Monocytes (2.5e5 per well) then were seeded in a 48-well plate with a final volume of 500 uL. Five days later (on day 5), fresh medium with cytokines were added to generate M2 macrophages. To generate primary human M2 macrophages, 250 pL of ImmunoCult™- SF Macrophage Medium containing 150 ng/mL human M-CSF + 60 ng/rnL human IL-4 + 60 ng/rnL human IL- 10 was added per well and incubated for 48 hours. Kupffer and HUVEC cells were seeded one day prior to infection, with the Kupffer cells seeded in Collagen-I treated wells and co-cultured with densely-seeded hepatocytes to promote apoptotic cell death. On day 7, duplicate wells were infected at an MOI of 100, for two hours in RPMI, with the YS I646Ar/.sc//AFLG/A/?^/-'/Aa//.s7/ Ac.s^'/J strains containing either a plasmid encoding NanoLuciferase® or huIL-15Ra-IL-15sc + huSTING N154S/R284G tazCTT. The cells then were washed 3 times with PBS and resuspended in RPMI + 100 pg/mL gentamicin (Sigma). After 48 hours, supernatants were harvested, cleared of cell debris and analysed for cytokine profile using a U- Plex® (Meso Scale Discovery).

As shown in the table below, along with the on-target, payload-mediated cytokines produced by the M2 macrophages treated with the YS I646A7/.sc/AFLG/A/?<:/g/ ^> /A<:7//.sA Ac..sg/) strain encoding huIL-15Ra-IL-15sc + huSTING N154S/R284G tazCTT, payload delivery also occurred in the Kupffer cells cultured with apoptotic hepatocytes. IFN-P and IP- 10, both secondary cytokines from STING activation, were induced in a payload-dependent manner in M2 macrophages and Kupffer + apoptotic hepatocytes, but not in the non-phagocytic HUVEC cells. These data indicate that Kupffer cells can become permissive to payload delivery when induced to proliferate with apoptotic hepatocytes.

EXAMPLE 55

Fibrotic Diseases Are Enriched in Proliferating Macrophages and Provides Delivery of Gene Therapies Encoding Therapeutic Payloads to Tissue-Resident

Macrophages

The finding that macrophages and liver Kupffer cells can be induced to proliferate by co-culturing them with apoptotic cells has important therapeutic implications. In the case of Kupffer cells, it has been described in the literature in the field of non-alcoholic steatohepatitis (NASH), a disease where liver damage leads to inflammation and liver fibrosis, that Kupffer cells phagocytose damaged hepatocytes (Li et al., (2020) Front Immunol 11:1169). Similarly, proliferating SPP1 ⁺ macrophages have been described in the diseased fibrotic lungs of IPF patients and not in healthy tissues, and they increase with disease progression (Morse etal., (2019) Eur Respir J 54(2): 1802441). As such, gene delivery of plasmids containing therapeutic payloads to these tissues results in the DNA being transferred and expressed by these proliferating macrophages, and payloads will not be expressed in other cells. This provides a therapeutic opportunity for treating lung and liver fibrosis patients with DNA-containing therapeutics.

EXAMPLE 56

Therapies That Promote Tumor Apoptosis Provide Enrichment of Phagocytic and Proliferating Macrophages and Enhance Payload Delivery of Gene Therapies in Tumor-Resident Macrophages

The ability to enhance plasmid payload delivery in tumors can be achieved through induction of apoptosis, which recruits phagocytic and proliferating macrophages that are required for DNA transfer to tumor-resident macrophages. Several types of chemotherapy have been described to promote tumor apoptosis, including docetaxel (DTX), paclitaxel (PTX), doxorubicin (DOX), 5 -fluorouracil (5- FU), carboplatin (CARB), and cyclophosphamide (CTX) (Anfray etal., (2019) Cells 9(1):46). Treatment of mice bearing 4T1 breast tumors with CTX, DTX, DOX and 5- FU all induced significant F4/80 ⁺ macrophage infiltration and enhanced disease progression (Zhao et al., (2017) Cancer Res 77(21):6021-6032). A similar effect was observed in CT26 colon cancer models, where treatment of tumor bearing mice, or tumor cells prior to implantation with 5-FU resulted in significant infiltration of SPP1 ⁺ macrophages and disease progression (Chang etal., (2019) FASEB J 33(1): 114-125). Similar studies could be clinically translated, where patients could be selected for that have been previously treated with apoptotic chemotherapies and have progressed. To this end, an analysis was performed on published data of the breast cancer PROMIX trial, where patients were assessed for biomarkers following either 2 cycles of epirubicin and docetaxol, or 4 cycle of chemotherapy + surgery and bevacizumab (Kimbung et al., (2018) Int J Cancer 142(3):618-628). The published datasets were then analyzed for tumor expression of CD68 following these treatments, as well as the expression of CD68 in each treatment cycle in the basal, HER2, and luminal A/B PAM50 subtypes.

As shown in the box plots in FIG. 28, significant increases in CD68 ⁺ macrophages were observed across all breast cancer patients after 2 cycles of chemotherapy with apoptosis-inducing DOX, while CD68 ⁺ macrophages were significantly reduced in patients treated for 4 cycles with chemotherapy + bevacizumab and surgery. The effect was even more pronounced for basal and luminal subtypes (see FIG. 29). These data show that combination therapy can be advantageous; the data show that prior treatment with chemotherapy can enhance the therapeutic effect of immunostimulatory bacteria provided here by increasing the number of proliferating macrophages in the tumors.

Other therapeutic strategies to enhance the amount of proliferating macrophages in tumors prior to dosing can be achieved with pre-treatment with anti- PD-1. The role that PD-1 and PD-L1 play on myeloid cell biology has been underappreciated relative to their roles in T-cell - tumor cell interactions (Lecoultre et al., (2020) JITC 8(2):e001408). In a previously published study using murine and human macrophages, PD-1 expression increased on tumor macrophages with increasing disease progression. PD-1 expression on macrophages was impaired phagocytosis, while treatment with anti-PD-1 reversed that effect (Gordon et al., (2017) Nature 545(7655):495-499). In a separate study, PD-L1 expression on macrophages was shown to be an indicator of macrophage activation. Treatment of macrophages with anti-PD-Ll repolarized them to an Ml -like inflammatory phenotype, and impaired phagocytosis (Hartley et al., Cancer Immunol Res 2020; Hartley et al., (2018) Cancer Immunol Res 6(10):1260-1273).

To determine the effects of the checkpoint therapies on the phagocytic functions of primary human macrophages, a study was performed using Ml and M2- derived primary human macrophages. For this, frozen human monocytes, isolated from healthy human donors, were thawed in complete medium (RPMI-1640 + 10% FBS), and washed by washed by centrifugation for 10 minutes at 600xg at room temperature. Monocytes were resuspended in ImmunoCult™- SF Macrophage Medium (StemCell Technologies) containing 100 ng/mL human M-CSF. Monocytes (2.5e5 per well) then were seeded in a 48-well plate with a final volume of 500 pL. Five days later (on day 5), fresh medium with cytokines were added to generate Ml and M2 macrophages. To generate primary human Ml macrophages, 250 pL of ImmunoCult™- SF Macrophage Medium containing 10 ng/mL LPS + 50 ng/mL FFNy was added per well and incubated for 48 hours. To generate primary human M2 macrophages, 250 pL of ImmunoCult™- SF Macrophage Medium containing 150 ng/mL human M-CSF + 60 ng/mL human IL-4 + 60 ng/mL human IL- 10 was added per well and incubated for 48 hours. 100-ug/mL of anti-mPD-Ll-mIgGle3 (InvivoFit) were added to the corresponding samples and incubated for 48 hours.

On day 7, duplicate wells were infected at an MOI of 10, for two hours in RPMI, with the following strains: strain containing a plasmid encoding huIL-15Ra-IL-15sc + huSTINGN154S/R284G tazCTT. The cells were washed 3 times with PBS and resuspended in RPMI + 100 pg/mL gentamicin (Sigma).

For flow cytometry analysis, cells were washed with 500-uL of DPBS and detached by incubating with 10-mM EDTA in PBS on ice for 20-min. Lifted cells were washed once with flow buffer (PBS + 2% FBS) and resuspended in 100-uL of LIVE/DEAD Violet Fixable Viability Dye. After 20-min incubation at RT, cells were washed once with flow buffer and resuspended in 50 pL of flow buffer containing True-Stain Monocyte Blocker and Human TruStain FcX. After 10-min incubation on ice, cell surface flow cytometry antibodies PD-L1 BV785 clone 2E9.2A3; PD-1 PE- Cy7 clone EH12.2H7 (all from BioLegend) were added. After 20-min of incubation on ice and in the dark, cells were washed twice with flow buffer. Cells then were resuspended in flow buffer and data were acquired using the NovoCyte® flow cytometer (ACEA Biosciences, Inc.) and analyzed using FlowJo™ software (Tree Star, Inc.).

For the phagocytosis assay, 1.5e5 CFSE-labeled apoptotic Raji (induced by 24 hours 1-mM H2O2 treatment) or pHrodo™ Green E. coli BioParticles™ Conjugate (Invitrogen) were added to each well and incubate at 37 °C. After 3 hours, the cells were washed with 500-uL of DPBS and detached by incubating with 10-mM EDTA in PBS on ice for 20-min. Lifted cells were washed once with flow buffer (PBS + 2% FBS) and resuspended in 50 pL of flow buffer containing True-Stain Monocyte Blocker and Human TruStain FcX. After 10-min incubation on ice, cell surface flow cytometry antibody SIRPa PE-Cy7 clone SE5A5 (BioLegend) were added. After 20- min of incubation on ice and in the dark, cells were washed twice with flow buffer. Cells then were resuspended in flow buffer and data were acquired using the NovoCyte® flow cytometer (ACEA Biosciences, Inc.) and analyzed using FlowJo™ software (Tree Star, Inc.). SIRPa-positive population was gated to differentiate macrophages from apoptotic Raji and pHrodo™ Green E. coli BioParticles™ Conjugate.

The results of this study confirm that treatment with anti-PD-Ll upregulates PD-1 expression and impairs macrophage phagocytosis of apoptotic tumor cells by M2 macrophages. Treatment of macrophages following infection with the immunostimulatory bacteria encoding huIL-15Ra-IL-15sc and huSTING N154S/R284G tazCTT with anti-PD-Ll restores the expression of PD-1 back to baseline. These data indicate a treatment protocol comprising: treatment with anti-PD-1, followed by the immunostimulatory bacteria, followed by immunotherapy, where patients are pre-treated with anti-PD-1 to suppress PD-1 expression on macrophages and promote their phagocytic capacity prior to treatment with the bacteria, followed by treatment with the immunostimulatory bacteria, and then delaying treatment with anti-PD-Ll until the plasmid payloads have been delivered so that PD-L 1 is then induced on macrophages, after which anti-PD-Ll therapy can be administered.

EXAMPLE 57

Exemplary Constructs and Plasmids For Immunostimulatory Bacteria

The delivery vehicles, particularly the immunostimulatory bacteria provided herein contain plasmids that encode a payload, such as a cytokine or cytokine and cytosolic DNA/RNA sensor, or any other combinations described or contemplated herein. Among the constructs included in the plasmids are those that encode a type I interferon (IFN), such as an IFNa and/or IFNb. The constructs are for inclusion in a plasmid in a immunostimulatory bacterium, including those with the genome modifications described throughout the disclosure herein, and in particular immuno stimulatory bacterium that have genome modifications resulting in attenuated

- 558 -

RECTIFIED SHEET (RULE 91) ISA/EP TLR2 activity or attenuated TLR2 and TLR4 and/or TLR5 activity. These include the bacteria that have genome modifications that result in penta-acyl ated LPS, lack flagella, and also lack curli fimbriae, such as, for example, the Salmonella bacteria with such genome modifications. For example, the bacteria as exemplified that have the phenotype YS I 646/1 c/ / ^SPGGIApagP/AansB/AcsgD/F-Apurl, or

Y S I 646zj z.sc// ISFGGIApagP/AansB/AcsgD/F-Apurl/AthyA, or any of the exemplified bacteria that accumulate in the tumor microenvironment, and particularly in tumorresident macrophages.

Encoding a type I interferon (IFN), such as an IFNa and/or INFb in such bacteria can result in conversion of an immune desert tumor or a T-cell excluded tumor into a “hot “tumor, which is responsive to immunotherapy, and also, can, depending upon the additional payload convert macrophage into the M1/M2 hybrid phenotype.

A series of exemplary constructs, including regulatory sequences are provided herein, and exemplary nucleic acid sequences are set forth in SEQ ID NOs. 502-501. The components are indicated in the sequence listing and described below. Protein sequences of encoded protein also are included in the sequence listing.

SEQ ID NOs:502-545 set forth the sequences of constructs containing human IFNa2 and/or IFN-b. The constructs include eukaryotic promoters and enhancers, such as viral promoters and enhancers, optional IRES sequences for translation in eukaryotic cells, peptide-encoding sequences, such as sequences that encode a 2A peptide for polycistronic sequences that result in two translated proteins, and other such regulatory sequences. Each of the constructs, as set forth in the sequence listing contains a kanR gene, which is optional and/or can be replaced or eliminated, for growth in selective medium. Other optional components can be included on the plasmids as described throughout the disclosure herein. The constructs can encode other polypeptides, including other immunostimulatory proteins, as described throughout the disclosure herein. The constructs can include spacers, LTRs, UTRs, origins of replication and other such sequences as known to those of skill in the art and/or as described throughout the disclosure herein.

SEQ ID NOs:549 and 551 set forth the nucleotide sequence of a human IFNa2 and human IFN-b, respectively. SEQ ID NOs: 550 and 552 set forth the amino acid sequence of human IFNa2 and human IFN-b, respectively. SEQ ID NO:502 sets forth the sequence of a pATI 1.76 (see Example 8, which describes the pATI 1.75 and 1.76 vectors)-based construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a TATA signal (nucleotides 2197-2203 and 3784-3790); a 5’ UTR (nucleotides 2212- 2255); a Kozak sequence (nucleotides 2256-2261 and 4161-4166); human IFNa2 opt 15 (NM_000605.4, nucleotides 2265-2825, where the start and stop codons are encoded by nucleotides 2262-2264 and 2826-2828, respectively); an HPRE (nucleotides 2838-3371); a bGH poly(A) signal (nucleotides 3380-3604); a promoter spacer (nucleotides 3605-3634); an EF-l-alpha promoter (nucleotides 3635-3846); a 5’ LTR (nucleotides 3859-4127); human IFN-b opt 15 (nucleotides 4170-4727, where the start and stop codons are encoded by nucleotides 4167-4169 and 4728-4730, respectively); a WPRE (nucleotides 4780-5386); an SV40 poly(A) signal (nucleotides 5391-5512); and KanR complement (nucleotides 5858-6667). SEQ ID NO:503 sets forth the sequence of a construct with the same components as SEQ ID NO: 502, except that human IFNa2 and IFN-b are not opt 15.

SEQ ID NO:504 sets forth the sequence of a pATI 1.76 construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a TATA signal (nucleotides 2197-2203); a 5’ UTR (nucleotides 2212- 2255); a Kozak sequence (nucleotides 2256-2261); human IFNa2 opt 1 (nucleotides 2265-2825, where the start codon is encoded by nucleotides 2262-2264); a GGS linker (nucleotides 2826-2834); T2A opt 1 (nucleotides 2835-2888); human IFN-b opt 1 (nucleotides 2889-3446, where the stop codon is encoded by nucleotides 3447- 3449); an HPRE (nucleotides 3460-3993); a bGH poly(A) signal (nucleotides 4002- 4226); a duplexed DNA insert (nucleotides 4205-4284); and KanR complement (nucleotides 4491-5300). SEQ ID NQ:505 sets forth the sequence of a construct with the same components as SEQ ID NO:504, except that human IFNa2, IFN-b and T2A are not opt 1.

SEQ ID NQ:506 sets forth the sequence of a pATI 1.76 construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a TATA signal (nucleotides 2197-2203); a 5’ UTR (nucleotides 2212- 2255); a Kozak sequence (nucleotides 2256-2261); human IFNa2 opt 12 (nucleotides 2265-2825, where the start and stop codons are encoded by nucleotides 2262-2264 and 2826-2828, respectively); a ribosomal binding site (VCIP IRES ATUM, nucleotides 2829-3418); human IFNb opt 12 (nucleotides 3422-3979, where the start and stop codons are encoded by nucleotides 3419-3421 and 3980-3982, respectively); an HPRE (nucleotides 3993-4526); a bGH poly(A) signal (nucleotides 4535-4759); a duplexed DNA insert (nucleotides 4738-4817); and KanR complement (nucleotides 5024-5833). SEQ ID NO:507 sets forth the sequence of a construct with the same components as SEQ ID NO:506, except that human IFNa2 and IFN-b are not opt 12.

SEQ ID NO:508 sets forth the sequence of a pATI 1.76 construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a TATA signal (nucleotides 2197-2203 and 3781-3787); a 5’ UTR (nucleotides 2212-2255); a Kozak sequence (nucleotides 2256-2261 and 4158-4163); human IFNb opt 13 (nucleotides 2265-2822, where the start and stop codons are encoded by nucleotides 2262-2264 and 2823-2825, respectively); an HPRE (nucleotides 2835-3368); a bGH poly(A) signal (nucleotides 3377-3601); a promoter spacer (nucleotides 3602-3631); an EF-1 alpha promoter (nucleotides 3632-3843); a 5’ LTR (nucleotides 3856-4124); human IFNa2 opt 13 (nucleotides 4167-4727, where the start and stop codons are encoded by nucleotides 4164-4166 and 4728-4730, respectively); a WPRE (nucleotides 4780-5368); a SV40 poly(A) signal (nucleotides 5391-5512); and KanR complement (nucleotides 5858-6667). SEQ ID NQ:509 sets forth the sequence of a construct with the same components as SEQ ID NO:508, except that human IFNa2 and IFN-b are not opt 13.

SEQ ID NQ:510 sets forth the sequence of a pATI 1.76 construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a TATA signal (nucleotides 2197-2203); a 5’ UTR (nucleotides 2212- 2255); a Kozak sequence (nucleotides 2256-2261); human IFNb opt 2 (nucleotides 2265-2822, where the start codon is encoded by nucleotides 2262-2264); a GGS linker opt 2 (nucleotides 2823-2831); a T2A opt 2 (nucleotides 2832-2885); human IFNa2 opt 2 (nucleotides 2886-3446, where the stop codon is encoded by nucleotides 3447-3449); an HPRE (nucleotides 3460-3993); a bGH poly(A) signal (nucleotides 4002-4226); a duplexed DNA insert (nucleotides 4205-4284); and KanR complement (nucleotides 4491-5300). SEQ ID NO:511 sets forth the sequence of a construct with the same components as SEQ ID NO:510, except that human IFNa2, IFN-b, the GGS linker and T2A are not opt 2.

SEQ ID NO 5 I 2 sets forth the sequence of a pATI 1.76 construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a TATA signal (nucleotides 2197-2203); a 5’ UTR (nucleotides 2212- 2255); a Kozak sequence (nucleotides 2256-2261); human IFNb opt 10 (nucleotides 2265-2822, where the start and stop codons are encoded by nucleotides 2262-2264 and 2823-2825, respectively); a ribosomal binding site (VCIP IRES ATUM, nucleotides 2826-3415); human IFNa2 opt 10 (nucleotides 3419-3979, where the start and stop codons are encoded by nucleotides 3416-3418 and 3980-3982, respectively); an HPRE (nucleotides 3993-4526); a bGH poly(A) signal (nucleotides 4535-4759); a duplexed DNA insert (nucleotides 4738-4817); and KanR complement (nucleotides 5024-5833). SEQ ID NO:513 sets forth the sequence of a construct with the same components as SEQ ID NO:512, except that human IFNa2 and IFN-b are not opt 10.

SEQ ID NOAM sets forth the sequence of a pATI 1.76 construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a TATA signal (nucleotides 2197-2203); a 5’ UTR (nucleotides 2212- 2255); a ribosomal binding site (VCIP IRES, nucleotides 2256-2802); human IFNa2 opt 8 (nucleotides 2806-3366, where the start codon is encoded by nucleotides 2803- 2805); a GGS linker opt 8 (nucleotides 3367-3375); T2A opt 8 (nucleotides 3376- 3429); human IFN-b opt 8 (nucleotides 3430-3987, where the stop codon is encoded by nucleotides 3988-3990); an HPRE (nucleotides 4001-4534); a bGH poly(A) signal (nucleotides 4543-4767); a duplexed DNA insert (nucleotides 4746-4825); and KanR complement (nucleotides 5032-5841). SEQ ID NO:515 sets forth the sequence of a construct with the same components as SEQ ID NO: 514, except that human IFNa2, IFN-b, the GGS linker and T2A are not opt 8.

SEQ ID NO: 516 sets forth the sequence of a pATI 1.76 construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a TATA signal (nucleotides 2197-2203); a 5’ UTR (nucleotides 2212- 2255); a ribosomal binding site (VCIP IRES, nucleotides 2256-2802); human IFNb opt 7 (nucleotides 2806-3363, where the start codon is encoded by nucleotides 2803- 2805); a GGS linker opt 7 (nucleotides 3364-3372); T2A opt 7 (nucleotides 3373- 3426); human IFNa2 opt 7 (nucleotides 3427-3987, where the stop codon is encoded by nucleotides 3988-3990); an HPRE (nucleotides 4001-4534); a bGH poly(A) signal (nucleotides 4543-4767); a duplexed DNA insert (nucleotides 4746-4825); and KanR complement (nucleotides 5032-5841). SEQ ID NO:517 sets forth the sequence of a construct with the same components as SEQ ID NO: 516, except that human IFNa2, IFN-b, the GGS linker and T2A are not opt 7.

SEQ ID NO:518 sets forth the sequence of a pATI 1.77 construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a TATA signal (nucleotides 2197-2203 and 3827-3833); a 5’ UTR (nucleotides 2212-2298); a Kozak sequence (nucleotides 2299-2304 and 4204-4209); human IFNa2 opt 40 (nucleotides 2308-2868, where the start and stop codons are encoded by nucleotides 2305-2307 and 2869-2871, respectively); an HPRE (nucleotides 2881-3414); a bGH poly(A) signal (nucleotides 3423-3647); a promoter spacer (nucleotides 3648-3677); an EF-l-alpha promoter (nucleotides 3678-3889); a 5’ LTR (nucleotides 3902-4170); human IFN-b opt 40 (nucleotides 4213-4770, where the start and stop codons are encoded by nucleotides 4210-4212 and 4771-4773, respectively); a WPRE (nucleotides 4823-5411); an SV40 poly(A) signal (nucleotides 5434-5555); and KanR complement (nucleotides 5901-6710). SEQ ID NO:519 sets forth the sequence of a construct with the same components as SEQ ID NO: 518, except that human IFNa2 and IFN-b are not opt 40.

SEQ ID NO:520 sets forth the sequence of a pATI 1.77 construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a TATA signal (nucleotides 2197-2203); a 5’ UTR (nucleotides 2212- 2298); a Kozak sequence (nucleotides 2299-2304); human IFNa2 opt 28 (nucleotides 2308-2868, where the start codon is encoded by nucleotides 2305-2307); a GGS linker opt 28 (nucleotides 2869-2877); T2A opt 28 (nucleotides 2878-2931); human IFN-b opt 28 (nucleotides 2932-3489, where the stop codon is encoded by nucleotides 3490-3492); an HPRE (nucleotides 3503-4036); a bGH poly(A) signal (nucleotides 4045-4269); a duplexed DNA insert (nucleotides 4248-4327); and KanR complement (nucleotides 4534-5343). SEQ ID NO:521 sets forth the sequence of a construct with the same components as SEQ ID NO: 520, except that human IFNa2, IFN-b, the GGS linker and T2A are not opt 28.

SEQ ID NO:522 sets forth the sequence of a pATI 1.77 construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a TATA signal (nucleotides 2197-2203); a 5’ UTR (nucleotides 2212- 2298); a Kozak sequence (nucleotides 2299-2304); human IFNa2 opt 28 (nucleotides 2308-2868, where the start codon is encoded by nucleotides 2305-2307); a GGS linker opt 28 (nucleotides 2869-2877); T2A opt 28 (nucleotides 2878-2931); human IFN-b opt 28 (nucleotides 2932-3489, where the stop codon is encoded by nucleotides 3490-3492); an HPRE (nucleotides 3503-4036); a SV40 poly(A) signal (nucleotides 4067-4188); and KanR complement (nucleotides 4395-5204). SEQ ID NO:523 sets forth the sequence of a construct with the same components as SEQ ID NO: 522, except that human IFNa2, IFN-b, the GGS linker and T2A are not opt 28.

SEQ ID NO:524 sets forth the sequence of a pATI 1.77 construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a TATA signal (nucleotides 2197-2203); a 5’ UTR (nucleotides 2212- 2298); a Kozak sequence (nucleotides 2299-2304); human IFNa2 opt 36 (nucleotides 2308-2868, where the start and stop codons are encoded by nucleotides 2305-2307 and 2869-2871, respectively); a ribosomal binding site (VCIP IRES ATUM, nucleotides 2872-3461); human IFN-b opt 36 (nucleotides 3465-4022, where the start and stop codons are encoded by nucleotides 3462-3464 and 4023-4025, respectively); an HPRE (nucleotides 4036-4569); a bGH poly(A) signal (nucleotides 4578-4802); a duplexed DNA insert (nucleotides 4781-4860); and KanR complement (nucleotides 5067-5876). SEQ ID NO:525 sets forth the sequence of a construct with the same components as SEQ ID NO:524, except that human IFNa2 and IFN-b are not opt 36.

SEQ ID NO:526 sets forth the sequence of a pATI 1.77 construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a TATA signal (nucleotides 2197-2203); a 5’ UTR (nucleotides 2212- 2298); a Kozak sequence (nucleotides 2299-2304); human IFNa2 opt 36 (nucleotides 2308-2868, where the start and stop codons are encoded by nucleotides 2305-2307 and 2869-2871, respectively); a ribosomal binding site (VCIP IRES ATUM, nucleotides 2872-3461); human IFN-b opt 36 (nucleotides 3465-4022, where the start and stop codons are encoded by nucleotides 3462-3464 and 4023-4025, respectively); an HPRE (nucleotides 4036-4569); a SV40 poly(A) signal (nucleotides 4600-4721); and KanR complement (nucleotides 4928-5737). SEQ ID NO: 527 sets forth the sequence of a construct with the same components as SEQ ID NO: 526, except that human IFNa2 and IFN-b are not opt 36.

SEQ ID NO:528 sets forth the sequence of a pATI 1.77 construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a TATA signal (nucleotides 2197-2203 and 3824-3830); a 5’ UTR (nucleotides 2212-2298); a Kozak sequence (nucleotides 2299-2304 and 4201-4206); human IFNb opt 41 (nucleotides 2308-2865, where the start and stop codons are encoded by nucleotides 2305-2307 and 2866-2868, respectively); an HPRE (nucleotides 2878-3411); a bGH poly(A) signal (nucleotides 3420-3644); a promoter spacer (nucleotides 3645-3674); an EF-1 alpha promoter (nucleotides 3675-3886); a 5’ LTR (nucleotides 3899-4167); human IFNa2 opt 41 (nucleotides 4210-4770, where the start and stop codons are encoded by nucleotides 4207-4209 and 4771-4773, respectively); a WPRE (nucleotides 4823-5411); a SV40 poly(A) signal (nucleotides 5434-5555); and KanR complement (nucleotides 5901-6710). SEQ ID NO:529 sets forth the sequence of a construct with the same components as SEQ ID NO:508, except that human IFNa2 and IFN-b are not opt 41.

SEQ ID NO:530 sets forth the sequence of a pATI 1.77 construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a TATA signal (nucleotides 2197-2203); a 5’ UTR (nucleotides 2212- 2298); a Kozak sequence (nucleotides 2299-2304); human IFNb opt 29 (nucleotides 2308-2865, where the start codon is encoded by nucleotides 2305-2307); a GGS linker opt 29 (nucleotides 2866-2874); T2A opt 29 (nucleotides 2875-2928); human IFNa2 opt 29 (nucleotides 2929-3489, where the stop codon is encoded by nucleotides 3490-3492); an HPRE (nucleotides 3503-4036); a bGH poly(A) signal (nucleotides 4045-4269); a duplexed DNA insert (nucleotides 4248-4327); and KanR complement (nucleotides 4534-5343). SEQ ID NO:531 sets forth the sequence of a construct with the same components as SEQ ID NO: 530, except that human IFNa2, IFN-b, the GGS linker and T2A are not opt 29. SEQ ID NO:532 sets forth the sequence of a pATI 1.77 construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a TATA signal (nucleotides 2197-2203); a 5’ UTR (nucleotides 2212- 2298); a Kozak sequence (nucleotides 2299-2304); human IFNb opt 29 (nucleotides 2308-2865, where the start codon is encoded by nucleotides 2305-2307); a GGS linker opt 29 (nucleotides 2866-2874); T2A opt 29 (nucleotides 2875-2928); human IFNa2 opt 29 (nucleotides 2929-3489, where the stop codon is encoded by nucleotides 3490-3492); an HPRE (nucleotides 3503-4036); a SV40 poly(A) signal (nucleotides 4067-4188); and KanR complement (nucleotides 4395-5204). SEQ ID NO: 533 sets forth the sequence of a construct with the same components as SEQ ID NO:532, except that human IFNa2, IFN-b, the GGS linker and T2A are not opt 29.

SEQ ID NO:534 sets forth the sequence of a pATI 1.77 construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a TATA signal (nucleotides 2197-2203); a 5’ UTR (nucleotides 2212- 2298); a Kozak sequence (nucleotides 2299-2304); human IFNb opt 37 (nucleotides 2308-2865, where the start and stop codons are encoded by nucleotides 2305-2307 and 2866-2868, respectively); a ribosomal binding site (VCIP IRES ATUM, nucleotides 2869-3458); human IFNa2 opt 37 (nucleotides 3462-4022, where the start and stop codons are encoded by nucleotides 3459-3461 and 4023-4025, respectively); an HPRE (nucleotides 4036-4569); a bGH poly(A) signal (nucleotides 4578-4802); a duplexed DNA insert (nucleotides 4781-4860); and KanR complement (nucleotides 5067-5876). SEQ ID NO:535 sets forth the sequence of a construct with the same components as SEQ ID NO:534, except that human IFNa2 and IFN-b are not opt 37.

SEQ ID NO:536 sets forth the sequence of a pATI 1.77 construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a TATA signal (nucleotides 2197-2203); a 5’ UTR (nucleotides 2212- 2298); a Kozak sequence (nucleotides 2299-2304); human IFNb opt 37 (nucleotides 2308-2865, where the start and stop codons are encoded by nucleotides 2305-2307 and 2866-2868, respectively); a ribosomal binding site (VCIP IRES ATUM, nucleotides 2869-3458); human IFNa2 opt 37 (nucleotides 3462-4022, where the start and stop codons are encoded by nucleotides 3459-3461 and 4023-4025, respectively); an HPRE (nucleotides 4036-4569); a SV40 poly(A) signal (nucleotides 4600-4721); and KanR complement (nucleotides 4928-5737). SEQ ID NO: 537 sets forth the sequence of a construct with the same components as SEQ ID NO 536, except that human IFNa2 and IFN-b are not opt 37.

SEQ ID NO:538 sets forth the sequence of a pATI 1.77 construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a TATA signal (nucleotides 2197-2203); a 5’ UTR (nucleotides 2212- 2298); a ribosomal binding site (VCIP IRES, nucleotides 2299-2845); human IFNa2 opt 35 (nucleotides 2849-3409, where the start codon is encoded by nucleotides 2846- 2848); a GGS linker opt 35 (nucleotides 3410-3418); T2A opt 35 (nucleotides 3419- 3472); human IFNb opt 35 (nucleotides 3473-4030, where the stop codon is encoded by nucleotides 4031-4033); an HPRE (nucleotides 4044-4577); a bGH poly(A) signal (nucleotides 4586-4810); a duplexed DNA insert (nucleotides 4789-4868); and KanR complement (nucleotides 5075-5884). SEQ ID NO:539 sets forth the sequence of a construct with the same components as SEQ ID NO: 538, except that human IFNa2, IFN-b, the GGS linker and T2A are not opt 35.

SEQ ID NO:540 sets forth the sequence of a pATI 1.77 construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a TATA signal (nucleotides 2197-2203); a 5’ UTR (nucleotides 2212- 2298); a ribosomal binding site (VCIP IRES, nucleotides 2299-2845); human IFNa2 opt 35 (nucleotides 2849-3409, where the start codon is encoded by nucleotides 2846- 2848); a GGS linker opt 35 (nucleotides 3410-3418); T2A opt 35 (nucleotides 3419- 3472); human IFNb opt 35 (nucleotides 3473-4030, where the stop codon is encoded by nucleotides 4031-4033); an HPRE (nucleotides 4044-4577); a SV40 poly(A) signal (nucleotides 4608-4729); and KanR complement (nucleotides 4936-5745). SEQ ID NO: 541 sets forth the sequence of a construct with the same components as SEQ ID NO: 540, except that human IFNa2, IFN-b, the GGS linker and T2A are not opt 35.

SEQ ID NO:542 sets forth the sequence of a pATI 1.77 construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a TATA signal (nucleotides 2197-2203); a 5’ UTR (nucleotides 2212- 2298); a ribosomal binding site (VCIP IRES, nucleotides 2299-2845); human IFNb opt 33 (nucleotides 2849-3406, where the start codon is encoded by nucleotides 2846- 2848); a GGS linker opt 33 (nucleotides 3407-3415); T2A opt 33 (nucleotides 3416- 3469); human IFNa2 opt 33 (nucleotides 3470-4030, where the stop codon is encoded by nucleotides 4031-4033); an HPRE (nucleotides 4044-4577); a bGH poly(A) signal (nucleotides 4586-4810); a duplexed DNA insert (nucleotides 4789-4868); and KanR complement (nucleotides 5075-5884). SEQ ID NO:543 sets forth the sequence of a construct with the same components as SEQ ID NO:542, except that human IFNa2, IFN-b, the GGS linker and T2A are not opt 33.

SEQ ID NO:544 sets forth the sequence of a pATI 1.77 construct that contains a high-copy number ColEl/pMBl/pBR322/pUC origin of replication (nucleotides 822-1410); a CMV enhancer (nucleotides 1704-2007); a CMV promoter (nucleotides 2008-2211); a TATA signal (nucleotides 2197-2203); a 5’ UTR (nucleotides 2212- 2298); a ribosomal binding site (VCIP IRES, nucleotides 2299-2845); human IFNb opt 33 (nucleotides 2849-3406, where the start codon is encoded by nucleotides 2846- 2848); a GGS linker opt 33 (nucleotides 3407-3415); T2A opt 33 (nucleotides 3416- 3469); human IFNa2 opt 33 (nucleotides 3470-4030, where the stop codon is encoded by nucleotides 4031-4033); an 1TPRE (nucleotides 4044-4577); a SV40 poly(A) signal (nucleotides 4608-4729); and KanR complement (nucleotides 4936-5745). SEQ ID NO: 545 sets forth the sequence of a construct with the same components as SEQ ID NO: 544, except that human IFNa2, IFN-b, the GGS linker and T2A are not opt 33.

EXAMPLE 58

THP1 Monocyte/Macrophage Cell Line Bactofection and Payload Expression Material & Methods

For Examples 58-60, the bacterial strain, unless otherwise noted, is the Salmonella strain that is SpurU SmsbBI Sa cl! SfljBI SfliC! SpagPI SansBI Scsgl) and contains a plasmid (SEQ ID NO.: 501) encoding the chimeric STING and IL-15 receptor complex. It is exemplary of STACT bacteria.

THP1 differentiation into macrophages (M0, Ml, M2)

Wild type (WT) Dual THP-1 cells were seeded in 96-well plate at le5 cells/well, in RPMI media supplemented with 10% FBS and 20ng/ml PMA (ThermoFisher, J 3916 MCR). PMA was washed away after 24h incubation and fresh complete RPMI media was added to rest for 3 days before polarization. M1/M2 polarization was done with treatment of 250ng/ml LPS + 25ng/ml IFNgamma or 20ng/ml IL-4 + 20ng/ml IL- 10 for 48h. Labeling of immunostimulatory bacteria with dyes

YS1646Atz5t//AFLG/A/2tz /A«n5 _J B/Ac5gD (YS1646 is ^purl! msbB) encoding the chimeric STING and IL-15 receptor complex on the plasmid, where the plasmid has the sequence set forth in SEQ ID NO: 501 (also referred to as a STACT bacterium in this example), was dyed with pHRodo® dye following manufacturer’s protocol (Invitrogen, P35357), and then with CellTrace® Violet (Invitrogen, C34571) before bactofection. Stock bacteria of this strain were washed in their culture medium (17000rpm, 5min). The bacteria pellet was resuspended in 1ml of staining buffer, then mixed with pHRodo® dye and incubated for 2h at room temperature, protected from light. Bacterial growth medium was added to recover bacteria and scavenge unreacted dye. Bacteria pellet was collected and resuspend in PBS, then mixed with CellTrace® Violet and incubated for 20min, at room temperature, protected from light. RPMI, 10% FBS was added and incubated for 5min to remove free dye in the solution. Bacteria pellet was collected after centrifugation and resuspend in RPMI, 10% FBS to allow CellTrace® reagent to undergo acetate hydrolysis before bactofection as 50ul of bacteria solution was added to corresponding wells followed by centrifugation at 500g/5min, and incubation at 37°C/lh.

Flow cytometry for phenotypic markers and dyes

Untreated and STACT infected THP-1 cells were stained with live/dead stain (Invitrogen, L34992), lOOul/well for 30min and washed before being stained with 50ul/well of Fc block (Biolegend, 422302) and Monocyte blocker (Biolegend, 426103) for 10min/4°C. Antibody mixture of CD40 (BV421), HLA-DR (BV570), CD206 (BV605), CD86 (BV650), CD209 (PE), HLA-ABC (PE-Dazzle) and CD 163 (PE-Cy7) was mixed in FACS buffer (PBS, 2% FBS) and added to samples with 50ul/well. Samples were washed twice between each staining step with lOOul/well of FACS buffer. Vybrant® DyeCyle was diluted 1 :4000 in FACS buffer and added to samples at lOOul/well right before proceeding to flow analysis with flow cytometer (NovoCyte® 3000) without washing.

STING reporter Luciferase assay

After bactofection of the cells, bacteria solution was removed and replaced with fresh complete RPMI media containing 50ug/ml Gentamicin. Supernatant was collected at 90min, 3h, 24h, and 48h after bactofection to detect luciferase by QUANTI-Luc™ (InvivoGen, rep-qlcl) and luminometer. Briefly, 25ul of supernatant sample was added to each well of opaque 96-well plate, then 50ul of QUANTI-Luc™ assay solution was added to each well and gently the plate was tapped to mix the solution and proceed immediately to luciferase measurement using luminometer.

Cell cycle assessment

Vybrant® DyeCycle (Invitrogen, V35005) was diluted 1 :4000 in FACS buffer and added to samples in 96-well (lOOul/well) before proceeding flow cytometer (NovoCyte® 3000) analysis without washing.

Results pHRodo® is a dye that is fluorescent only in acidic cellular compartments, such as in late endosomes and lysosomes. Salmonella is known to facilitate its uptake into cells and reside in phagosomes (F Garcia-del Portillo (1996) Microbiologica pubmed.ncbi.nlm.nih.gov/21178464/doi: 10.3791/56514).

FIG. 30 shows pHRodo® and CellTrace® labeling of YS I646Ac/.stZAFLG/A/x/g/ ^> /Ar///.sA/Ac ^, .sg/J containing the plasmid encoding IL-15 receptor complex and chimeric STING, where the plasmid has the sequence set forth in SEQ ID NO:501, over time internalized by THP1 differentiated into M0. Wild type (WT) Dual THP-1 cells were seeded in 96-well plate at le5 cells/well, in RPMI media supplemented with 10% FBS and 20ng/ml PMA. PMA was washed away after 24h incubation and fresh complete RPMI media was added to rest for 3 days before bactofection. YS1646Aa%7/AFLG/A/?czgP/A6znsB/Ac.sg£> containing the plasmid encoding IL-15 receptor complex and chimeric STING, where the plasmid has the sequence set forth in SEQ ID NO: 501, was labeled with pHRodo® dye following manufacturer’s protocol (Invitrogen, P35357) and then CellTrace® Violet (Invitrogen, C34571) before bactofection at MOI 1, 5, 20, and 40. Briefly, 50ul of bacteria solution was added to corresponding wells followed by centrifugation at 500g/5min, and incubation at 37°C/lh. Bacteria solution was removed and fresh complete RPMI media containing 50ug/ml Gentamicin was added until cells were harvested for flow analysis. Cells were stained with live/dead stain (Invitrogen, L34992) for 30min before being analyzed by flow cytometer (NovoCyte® 3000). pHRodo positive cells correlate with MOI increase at 90min to 48hrs (FIG. 30). pHRodo positive cells increase from 90min to 48hrs indicating increased bacterial presence within lysosomes.

As shown in FIGs. 31A-F, M0, Ml, and M2 macrophages show uptake of YS1646Aa%//AFLG/A/?^gP/Aans _J 8/AcsgD containing the plasmid encoding IL-15 receptor complex and chimeric STING, where the plasmid has the sequence set forth in SEQ ID NO:501, internally. Wild type (WT) Dual THP-1 cells were seeded in 96- well plate at le5 cells/well, in RPMI media supplemented with 10% FBS and 20ng/ml PMA. PMA was washed away after 24h incubation and fresh complete RPMI media was added to the rest for 3 days before polarization. M1/M2 polarization was done with treatment of 250ng/ml LPS + 25ng/ml ZFNy or 20ng/ml IL-4 + 20ng/ml IL- 10 for 48h. The strain STACT containing a plasmid (SEQ ID NO:501) encoding a chimeric constitutive STING and IL-15/IL-15R alpha chain complex was dyed with pHRodo® dye following manufacturer’s protocol (Invitrogen, P35357) and then CellTrace® Violet (Invitrogen, C34571) before bactofection at MOI 1, 5, 20, and 40. 50ul of bacteria solution was added to corresponding wells followed by centrifugation at 500g/5min, and incubation at 37°C/lh. Bacteria solution was removed and fresh complete RPMI media containing 50ug/ml Gentamicin was added until cells were harvested for flow analysis. Data were analyzed by FlowJo 10 and graphed by Prism 9.

The bacteria are internalized and trafficked to acidic lysosomes and nuclei in human macrophages (FIGs. 31A-F). M0, Ml and M2 macrophages exhibit increased intracellular presence and lysosomal trafficking of YS I646A€/.st/ FLG/A/?€/ _i g/7Ar///.sA/Ac.s /J containing the plasmid encoding IL-15 receptor complex and chimeric STING, where the plasmid has the sequence set forth in SEQ ID NO:501, with increasing MOI (FIGs. 31A-F).

FIGs. 32A and 32B depict STING reporter luciferase activity in wild-type M0 THP1 derived macrophages. Wild type (WT) Dual THP-1 cells were seeded in 96- well plate at le5 cells/well, in RPMI media supplemented with 10% FBS and 20ng/ml PMA. PMA was washed away after 24h incubation and fresh complete RPMI media was added to rest for 3 days before bactofection.

YS1646Aa5t//AFLG/A/?a P/A«n5B/Ac5gD containing the plasmid encoding IL- 15 receptor complex and chimeric STING, where the plasmid has the sequence set forth in SEQ ID NO:501, was dyed with pHRodo® dye following manufacturer’s protocol (Invitrogen, P35357) and then CellTrace® Violet (Invitrogen, C34571) before bactofection at MOI 1, 5, 20, and 40. Briefly, 50ul of bacteria solution was added to corresponding wells followed by centrifugation at 500g/5min, and incubation at 37C/lh. Bacteria solution was removed and replaced with fresh complete RPMI media containing 50ug/ml Gentamicin. cGAMP (InvivoGen, tlrl- nacga23-02) was added at 20ng/ml. Supernatant was collected at 90min, 3h, 24, and 48h after bactofection to detect luciferase by QUANTI-Luc™ (InvivoGen, rep-qlcl) and luminometer.

Increased STING reporter luciferase expression is observed with increasing YS I646Ac/.s<7 FLG/A/)c/ /7Ar///.sA/Ac ^, .s /J containing the plasmid encoding IL-15 receptor complex and chimeric STING, where the plasmid has the sequence set forth in SEQ ID NO:501, MOI (FIGs. 32A and 32B) and pHRodo positivity at 90min and 48hrs (FIG. 30).

FIG. 33 depicts cell cycle dependent internalization of STACT. Wild type (WT) Dual THP-1 cells were seeded in a 96-well plate at le5 cells/well, in RPMI medium supplemented with 10% FBS and 20ng/ml PMA. PMA was washed away after 24h incubation and fresh complete RPMI medium was added to rest for 3 days before polarization. M1/M2 polarization was done with treatment of 250ng/ml LPS + 25ng/ml IFNy or 20ng/ml IL-4 + 20ng/ml IL- 10 for 48h.

YS 1646A< .st/ ^/ AFLG/ pagPI ansBI csgl) containing the plasmid encoding IL-15 receptor complex and chimeric STING, where the plasmid has the sequence set forth in SEQ ID NO:501, was dyed with pHRodo® dye following manufacturer’s protocol (Invitrogen, P35357) before bactofection with the plasmid-containing STACT, and also with the STACT bacterium without the plasmid as a control. 50ul of bacteria solution was added to corresponding wells followed by centrifugation at 500g/5min, and incubation at 37C/lh. Bacteria solution was removed and fresh complete RPMI media containing 50ug/ml Gentamicin was added until cells were harvested for flow analysis. Cells were stained with live/dead stain (Invitrogen, L34992) for 30min. Vybrant® DyeCycle (Invitrogen, V35005) was added right before proceeding flow cytometer (NovoCyte® 3000) analysis without washing. Data was analyzed by Flowlo 10 and graphed by Prism 9.

To quantify cell cycle phases, Vybrant® Dyecycle dye was used in live cells. As the dye increases, it delineates G0/G1 (low uptake of the dye) with intermediate uptake of the dye in cells in S phase and highest uptake of the dye observed in cells in the G2/M phase (FIG. 33, top left schematic panel). At 48hrs, the highest level of phRodo+ bacteria are seen in G2/M phase of cell cycle in macrophages (M0, Ml and M2), with a much smaller percentage of bacterial uptake in G0/G1 phase. These phenomena are independent of the differentiation phenotype of the macrophages (M0, Ml, M2) (FIG. 33). The results show that YS1646Aa.s<7/AFLG/A/x/g/VA<:w/.s7vAc'.sg/9 containing the plasmid encoding IL- 15 receptor complex and chimeric STING, where the plasmid has the sequence set forth in SEQ ID NO: 501, induces phenotypic changes in macrophages upon bactofection in cells that have the baceterium in lysosomes (z.e., pHRodo positive). YS1646Aasz// AFLG/ kpagPI kansBI \csgD containing the plasmid encoding IL-15 receptor complex and chimeric STING, where the plasmid has the sequence set forth in SEQ ID NO: 501, taken up by MO macrophages into lysosomes results in increased expression of Ml (HLA-DR, CD86) and M2 (CD163, CD206, CD209) markers measured via specific antibodies for each marker using flow cytometry at 90mins. The results also show that cell cycle internalization is independent of macrophage type, as the internalization of the bacterium that lacks a plasmid and encoded payload also is cell cycle dependent.

EXAMPLE 59

Anti-PD-1 and Immunostimulatory Bacteria Combination Therapy Material & Methods

Cell Culture Growth and Harvest:

Cell Line: EMT6-M

Media: Waymouth’s and 15% Fetal Bovine Serum

Growth Conditions: 37°C, 5% CO2

Seed Cell Density: 3X10 ⁶

Implant Cell Density: lX10 ⁵ /Mouse

Implant Site: 2 ^nd Mammary Fat Pad, Orthotopic Implantation

Needle Gauge: 30G

Implant Volume: lOOpL/mouse

Mice:

Sex: Female

Age: 6-8 weeks

Source Vendor: Charles River

Bedding: Wood Chips

Enrichment: Paper Shred

Light/Dark Cycle: 12hr/12hr

Cell Implantation Methods

Cells are grown in culture flasks on site at Actym Therapeutics in Berkeley, CA, and prepared for injection once needed number of cells is obtained (i.e., 2-1 days). Cells are prepared to the correct cell density required and kept at 4°C (wet ice) while transferred via vehicle to Explora Marina Village in Alameda, CA. A ImL syringe is filled with ~0.6mL of the cell mixture The hub of the syringe is cleaned and dried with gauze. A 30G needle is attached to the filled syringe and the bevel is made to face up aligned with the volume indication numbers on the syringe. The needle and syringe are cleared of air bubbles and set to 0.5mL. A total of 5 mice (1 cage) are implanted using the same needle and syringe set up. Between animals the syringe is kept horizontal on ice on a rocking platform. Animals are implanted after a minimum of 3 days acclimatization time in the vivarium holding room. At 2.5-3% Isoflurane the mice are anesthetized in an induction chamber, the animal is transferred to the nose cone and placed supine on a clean absorbent bench pad. The second mammary fat pad is identified and shaved with clippers to remove fur from the immediate area. The shaved area is cleaned with a 70% isopropyl alcohol swab. Using fine pointed forceps, the area immediately adjacent to the nipple is held to isolate the nipple area. At time of implantation while gently lifting the said area the needle is introduced into the nipple area just under the skin. A total volume of 0. ImL (lOOmL) of cell mixture is injected to create a “bubbled up” appearance at the mammary fat pad. While using the fine forceps to pinch around the needle it is then removed. The skin is held in the forceps for 3 seconds after removal of the needle this to prevent leakage from the injection site. The area is observed for outflow and gently dried with gauze if necessary. The animal is removed from the nose cone and returned to the home cage for recovery. A new needle and syringe are used with every cage. The animals are visually monitored one day after cell implantation for abnormal behavior or adverse health.

Tumor Measurement and Volume Calculation: When tumors are palpable, generally, Day 4-6 post-implant (100,000 cell/mouse at implant) they are measured using Fowler Calipers and measurements are recorded and calculated using StudyLog® software. Specifically, tumor volumes are calculated using the following mathematical formula (Length X Width ² )/2 = mm ³ , all groups are randomized on the day of treatment or one day prior (as required by schedule constraints).

Group Randomization: Animals dosed at the schedule of 2-3 days prior to the primary therapy of YS I646A /.S Z/ AFLG/ pagPI ansBI csgD containing the plasmid encoding IL-15 receptor complex and chimeric STING, where the plasmid has the sequence set forth in SEQ ID NO:501, are normalized into groups based on body weight. For all other groups when tumors are a relevant size as determined by the study design (i.e., 50-75mm ³ ) the animals are randomized into groups based on a normalized tumor size. StudyLog® software randomizes the animals based on tumor volume accordingly. The animals are physically rearranged into cages with animals of the same group. Up to five animals are housed in a cage but no less than 2 animals are placed per cage i.e., no single housed animals are allowed).

Dilution of anti-PD-1 antibody:

Source:

InvivoGen Recombinant mouse mAb against murine PD-1, Anti-mPD-1- mIgGle3 (InvivoFit). After reconstitution of the lOmg lyophilized antibody with 5mL of sterile water the concentration is 2mg/mL. In a sterile 0.9% Sodium Chloride vial containing 7.5mL, a volume of 2.5mL of the anti-PD-1 (5mg) is added making a 0.5mg/mL solution. The solution is kept a 4°C for a maximum of 1 month.

Dosing anti-PD-1 antibody:

Each mouse is restrained, and the ventral portion is exposed. The animal is dosed into the intraperitoneal cavity at a volume of 0.2mL of the 0.5mg/mL using a 27G needle. The dosed amount is lOOpg/mouse. The dosing regimen is once every 3- 4 days i.e., 2X/week IP dosing.

Primary Therapy Dosing:

YS 646Aasd/ AFLG/ApagP/ ansB/ csgD containing the plasmid encoding IL- 15 receptor complex and chimeric STING, where the plasmid has the sequence set forth in SEQ ID NO:501, is dosed via the lateral tail vein of the mouse at a volume of 0.2mL/mouse in a single dose. The mouse is placed a restraint device after warming in a heated chamber. The area of injection is cleaned with a 70% isopropyl alcohol swab. A 27G needle is attached to a ImL syringe containing

YS Ibdb c/.s't// AFLG/ YpagPI ansB/ csgl) (containing the plasmid encoding IL-15 receptor complex and chimeric STING, where the plasmid has the sequence set forth in SEQ ID NO:501) with the appropriate CFU concentration. The needle is introduced into a lateral tail vein and the dosing solution is dosed at a volume of 0.2mL. The area is covered and pressured is applied with a clean gauze to stop any back flow of dosing solution and bleeding. The animal is placed in a recovery cage and allowed to return to normal body temperature while being monitored for any abnormal reactions to the manipulation. After complete recovery, the animal is placed in the home cage and returned to the holding rack. EMT6 typically is an immune excluded anti-PD-1 non-responsive breast cancer syngeneic orthotopic tumor model. Pre-treatment of EMT6 orthotopic mammary fat pad tumor bearing female balb/c mice with an anti-PD-1 antibody when dosed intraperitoneally two days prior (and every 3-4 days thereafter) to a single intravenous dose of YS1646 Aasd/i\FLG/^pagP/^ansB/^csgD containing the plasmid encoding IL-15 receptor complex and chimeric STING, where the plasmid has the sequence set forth in SEQ ID NO: 501, at 3e7 CFU. A lower than maximally efficient dose of YS 1646Ac7.vc//AFLG/A/?ag/7Ac///.s7YAc'.sg/) containing the plasmid encoding IL- 15 receptor complex and chimeric STING, where the plasmid has the sequence set forth in SEQ ID NO:501, was chosen in order to observe a combination effect. YS 1646A<7.st/ ^/ AFLG/ kpagP/ kansBI \csgD containing the plasmid encoding IL-15 receptor complex and chimeric STING, where the plasmid has the sequence set forth in SEQ ID NO:501, had a TGI of 30% compared to PBS group, with 29% cure rate. Anti-PD-1 monotherapy had a TGI of 0%, with no cures in the treatment group. In combination, YS1646Aa«7/AFLG/A/?czgP/A<2nsB/AcsgZ) containing the plasmid encoding IL-15 receptor complex and chimeric STING, where the plasmid has the sequence set forth in SEQ ID NO: 501, and anti-PD-1 mAb treatment had a TGI of 62% and 46% cure rate, when the antibody was provided 2 days prior to the treatment with YS1646Aasz// AFLG/ kpagPI kansBI \csgl) containing the plasmid encoding IL- 15 receptor complex and chimeric STING, where the plasmid has the sequence set forth in SEQ ID NO:501. When the anti-PD-1 was provided on the same day as YS I646A«.stZ ^/ AFLG/A/?«g/ ^> /At/».sA/Ac.sg/.) containing the plasmid encoding IL-15 receptor complex and chimeric STING, where the plasmid has the sequence set forth in SEQ ID NO:501, 13% tumor growth inhibition (TGI) and 7% cure rate was observed. Thus, pre-treatment with the anti-PD-1 antibody two (2) days prior to treatment with the immunostimulatory bacteria was more effective then pre-treatment one day prior to treatment.

EXAMPLE 60

Indication Expansion And Syngeneic Model Selection Analysis

Material & Methods

Dataset pre-processing

The Cancer Genome Atlas (TCGA) was downloaded from GDS Portal (portal.gdc.cancer.gov) and UCSD Xenia database (xenabrowser.net). MET500 data were downloaded from UCSC Xena database (xenabrowser.net). A syngeneic tumor model dataset was downloaded from TISMO database (tismo.cistrome.org).

Data Analysis

Since each gene has multiple transcripts with distinct ensemble IDs, ensemble ID level data were collapsed into gene-level log2 FPKM counts using collapseRows function in WGCNA package in R. Pathway signature scores were calculated as scaled z-scores across the genes within a given gene signature shown below. Heatmap and other plots were plotted using the pheatmap and ggplot2 packages respectively.

• Adenosine - CD39, CD73, AIR, A2a/bR, A3R

• Corvus ADO - IL1B, PTGS2, CXCL1, CXCL2, CXCL3, CXCL5, CXCL6, CXCL8 - Fong etal. (2020) Can Discovery doi: 10.1158/2159-8290.CD-19- 0980

• AZ ADO - PPARG, CYBB, COL3A1, FOXP3, LAG3, APP, CD81, GPI, PTGS2, CASP1, FOS, MAPK1, MAPK3, CREB1 - Sidders etal. (2020) Clin. Cancer Res. doi: 10.1158/1078-0432.CCR-19-2183

• Purine metabolism - PPAT, DCK, ATIC, IMPDH1, RRM2

(doi. org/10.3389/fonc.2020.583053)

• Antigen presentation

MHC I (HLA-A/B/C), MHC II (HLA-DP/DQ/DR)

• Krummel T cell infiltration (doi.org/10.1016/j. cell.2021.12.004)

• Krummel Stromal infiltration

• Krummel Myeloid infiltration • SPP1 and C1QC TAM (doi: 10.3389/fimmu.2021.694801)

• Glycolysis - GLUT 1, LDHA

• Hypoxia - HIFla, VEGFA, GLUT1

• Glutamine - GLS2

• TGFbeta - TGFB 1 , TGFBR 1 /2 Tumor indications are denoted by abbreviations shown in the Table below.

Study i i.i. . .. Study Name Abbreviation

LAML Acute Myeloid Leukemia

ACC Adrenocortical carcinoma

BLCA Bladder Urothelial Carcinoma

LGG Brain Lower Grade Glioma

BRCA Breast invasive carcinoma

CESC Cervical squamous cell carcinoma and endocervical adenocarcinoma

CHOL Cholangiocarcinoma

LCML Chronic Myelogenous Leukemia

COAD Colon adenocarcinoma

ESCA Esophageal carcinoma

GBM Glioblastoma multiforme

HNSC Head and Neck squamous cell carcinoma

KICK Kidney Chromophobe

KIRC Kidney renal clear cell carcinoma

KIRP Kidney renal papillary cell carcinoma

LIHC Liver hepatocellular carcinoma

LUAD Lung adenocarcinoma

LUSC Lung squamous cell carcinoma

DLBC Lymphoid Neoplasm Diffuse Large B-cell Lymphoma

MESO Mesothelioma

MISC Miscellaneous

OV Ovarian serous cystadenocarcinoma

PAAD Pancreatic adenocarcinoma

PCPG Pheochromocytoma and Paraganglioma

PRAD Prostate adenocarcinoma

READ Rectum adenocarcinoma

SARC Sarcoma A LL • Study Name Abbreviation

SKCM Skin Cutaneous Melanoma

ST AD Stomach adenocarcinoma

TGCT Testicular Germ Cell Tumors

THYM Thymoma

THCA Thyroid carcinoma

UCS Uterine Carcinosarcoma

UCEC Uterine Corpus Endometrial Carcinoma

UVM Uveal Melanoma

Results

As tumors grow in size, they tend to become hypoxic with high levels of HIFla (FIG. 34). As a consequence, several oncogenes such as TGFb, c-Myc coupled with necrosis and apoptosis are upregulated in the tumor microenvironment together leading to an increased level of ATP. This increase subsequently triggers CD39 and CD73, ectonucleotidases that break down ATP into AMP and adenosine respectively. STACT is dependent on adenosine as an auxotroph for its colonization and survival in the TME. It is difficult to assess tumor adenosine levels since adenosine tends to be quite unstable and taken up by cells. Therefore CD73, CD39 published adenosine signatures as surrogate markers of tumors with high adenosine that are coupled with high levels of hypoxia and TGFbeta pathways are used. FIG. 34 shows that high levels of adenosine and immune exclusion are triggered by hypoxia in tumors.

FIGs. 35A and 35B depict CD73 and CD39 expression across solid tumor indications in TCGA. Tumors with high CD73 and CD39 expression are shown in FIGs. 35A and 35B. In FIG. 35B, median expression of each tumor indication for CD73 and CD39 was obtained. This identifies top adenosine rich indications such as thyroid, NSCLC, ccRCC (KIRC), mesothelioma (MESO), colorectal (COAD, READ), esophageal (ESCA), stomach (STAD), sarcoma (SARC), pancreatic (PAAD), HNSC, liver (HCC) and glioma (LGG, GBM) cancers.

FIGs. 36A and 36B depict T cell, stromal and myeloid cell signatures across solid tumor indications in TCGA. Expression of T cell, myeloid and stromal signatures from Combes etal. (2020) Cell doi.org/10.1016/j . cell.2021.12.004 was used to identify tumors with high T cells, myeloid and stromal cell presence. This showed that indications such as thymoma, TGCT, ccRCC (KIRC), DLBCL, NSCLC (LU AD, LUSC), Mesothelioma (MESO), pancreatic (PAAD), breast, sarcoma and stomach cancer had the highest presence of T cells, myeloid and stromal cells (FIGs. 36A and 36B).

FIGs. 37A-F depict myeloid cell markers across Thorsson et al. (2018) defined immune subtypes C1-C6 in TCGA. Highest phagocytic myeloid population was observed in C6 Immune subtype (TGFb dominant) defined in Thorrson etal. (2018) Immunity doi.org/10 016/jjmmuni.2018.03.023 (FIGs. 37C-F). This also shows the connection between high TGFbeta, adenosine and high myeloid presence with an immunosuppressive phenotype since C6 subtype was shown to have highest levels of M2 macrophages (Thorsson et al. (2018) Immunity doi.org/10.1016/j.immuni.2018.03.023).

All of the above information was combined with previously published adenosine response signatures and metabolic pathways that typically associate with high levels of hypoxia. From this, top indications that associate with high colonization of immunostimulatory bacteria provided herein, such as the bacteria that lack flagella, are msbB'lpagP~, and that are auxotrophic for adenosine and derivatives thereof, were identified. Also identified were top indications that associate with high colonization of these bacteria due to high hypoxia, adenosine, myeloid and T cell presence and efficacy in primary tumors in TCGA (FIG. 38) and metastatic cancers in MET500 (FIG. 39). Both datasets identified top indications indicated in FIG. 38 and FIG. 39. FIG. 38 depicts indications with high myeloid, T cells, hypoxia, adenosine, TGFbeta and antigen presentation in TCGA. FIG. 39 depicts indications with high myeloid, T cells, hypoxia, adenosine, TGFbeta and antigen presentation in metastatic cancers in MET500 dataset.

FIGs. 40A-C depict syngeneic mouse models with high T cell, myeloid and stromal infiltration and high adenosine. T cell, myeloid, and stromal cell signatures in the syngeneic model RNAseq TISMO database were assessed using only naive untreated syngeneic in vivo tumors (FIGs. 40A-C). The expression per model was averaged across the technical or biological replicates. A ranking based scoring was used in this case where a set of genes within a T cell, myeloid and stromal signature was used to generate an average rank based on the ranking of individual genes.

As the myeloid cell infiltration increases, treatment with the immunostimulatory bacteria provided herein result in higher tumor colonization and anti-tumor efficacy. The high myeloid content (shown by CD68) is coupled with these models also exhibiting high adenosine in the TME given by expression of CD73 i.e., NT5E (FIG. 40C).

Solid tumor indications with high hypoxia, and TGFbeta promote high adenosine microenvironment and myeloid content. These indications will exhibit high colonization of the immunostimulatory bacteria, such as the STACT bacteria, such as those with a phenotype \piirll hmsbB/ Vasili kfljB! \fliCI kpagPI kansBI \csgD, in the tumor microenvironment as well as high level of uptake by myeloid cells. This behavior is exhibited in primary as well as metastatic tumors. Tumors include, for example chronic lymphoblastic leukemias (CLLs) and myeloid malignancies. CLL has high adenosine and a hypoxic tumor microenvironment, and myeloid malignancies because they are myeloid cells, which as shown and described herein are infected with the immunostimulatory bacteria provided herein.

Since modifications will be apparent to those of skill in the art, it is intended that this invention be limited only by the scope of the appended claims.